HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS		QUICK SEARCH:   [advanced] Author: Keyword(s): Year:  Vol:  Page:

This Article Abstract Full Text (PDF) Alert me when this article is cited Alert me if a correction is posted Services Similar articles in this journal Similar articles in PubMed Alert me to new issues of the journal Download to citation manager Citing Articles Citing Articles via Google Scholar Google Scholar Articles by Ugwumadu, A. Articles by Jeffrey, I. Search for Related Content PubMed PubMed Citation Articles by Ugwumadu, A. Articles by Jeffrey, I. Related Collections General obstetrics Infectious disease Obstetric complications of pregnancy Pathology Placenta

Oral Clindamycin and Histologic Chorioamnionitis in Women With Abnormal Vaginal Flora

From the Departments of ¹Obstetrics and Gynecology, ²Community Health Sciences, ³Genitourinary Medicine, and ⁴Cellular Pathology, St. George’s Hospital; and ⁵St. George’s University of London, London, United Kingdom.

	ABSTRACT

TOP ABSTRACT PATIENTS AND METHODS RESULTS DISCUSSION REFERENCES

 
OBJECTIVE: Oral clindamycin reduced late miscarriage and preterm birth in asymptomatic women with bacterial vaginosis or intermediate flora. We investigated whether clindamycin reduced the incidence of histologic chorioamnionitis as a mechanism for these beneficial effects.

METHODS: This was a subanalysis of 126 participants from a larger randomized controlled trial. We compared the incidence of histologic chorioamnionitis between the clindamycin and placebo groups. Histologic chorioamnionitis was diagnosed by the presence of polymorphonuclear leukocytes, separately in the amnion and chorion, decidua, fetal surface of the placenta, the walls of fetal chorionic vessels, umbilical cord, or in the subchorionic fibrin layer. Microbiologic cultures were done on swabs from the space between the chorion and amnion layers.

RESULTS: Histopathologic results were available for 122 placentas, 62 (51%) and 60 (49%) in the clindamycin and placebo groups, respectively. There were no significant differences in inflammation between the groups in the decidua (41% compared with 43%), membranes (25% compared with 41%), fetal vessels (16% compared with 14%), or subchorionic fibrin (32% compared with 34%). Adjusting for gestational age, ethnic origin, or history of miscarriage did not alter the results. There were no significant differences in the outcomes of pregnancy between women with and without inflammation, either before or after adjustment for treatment group.

CONCLUSION: Although oral clindamycin reduced late miscarriage and preterm birth in women with abnormal vaginal flora, this effect is unlikely to be mediated through a reduction in the incidence of histologic chorioamnionitis. The relatively small size of the groups, however, does not allow us to rule out a real effect, especially given the lower rate of membrane inflammation observed in the clindamycin group.

LEVEL OF EVIDENCE: I

Histologic chorioamnionitis has been shown to be more predictive of abnormal neonatal cerebral outcomes (periventricular echodensity or echolucency, ventriculomegaly, intraventricular hemorrhage, and seizures)⁵ than clinical chorioamnionitis, defined as maternal fever and the presence of any or more of the following: maternal or fetal tachycardia, uterine tenderness, foul smelling or purulent vaginal discharge, maternal leukocytosis, or raised C-reactive protein. Only 10–15% of histologic chorioamnionitis is recognized clinically.⁶ At present there is no evidence that any antibiotic is effective in treating histologic or clinical chorioamnionitis in any population of pregnant women, including those with abnormal vaginal flora. Therefore, the current management of clinical chorioamnionitis is the expeditious delivery of the fetus regardless of the gestational age at diagnosis. This may, however, increase the risk of prematurity-related morbidity and mortality. We recently reported that oral clindamycin therapy administered early to pregnant women with abnormal vaginal flora significantly reduced the incidence of second trimester miscarriage and preterm birth.⁷ The aim of this secondary analysis is to test the hypothesis that oral clindamycin reduced the incidence of histologic chorioamnionitis as a mechanism for the reduction of the risks of late miscarriage and preterm birth.

	PATIENTS AND METHODS

TOP ABSTRACT PATIENTS AND METHODS RESULTS DISCUSSION REFERENCES

 
This study was a subanalysis of 126 participants enrolled in a larger randomized controlled trial of the effect of oral clindamycin on the incidence of late miscarriage and preterm delivery in pregnant women with asymptomatic abnormal vaginal flora.⁷ Briefly, all pregnant women attending their first antenatal visit at St. George’s Hospital, London or St. Helier Hospital, Surrey from November 1, 1996, through to February 1, 1999, were offered screening for abnormal vaginal flora if they were between 12 and 22 weeks (mean 15.6 weeks) of pregnancy according to their menstrual date or early ultrasound scan. If they had bacterial vaginosis, Nugent score 7–10; or intermediate flora, Nugent score 4–6, met the study’s inclusion criteria, and consented to participate, they were randomly assigned to a treatment group (oral clindamycin 300 mg twice daily for 5 days or identical placebo) within 10 days of screening. We used a computer program to randomly assign the numbers 1 to 500 to clindamycin or placebo treatment. A trial pharmacist used this randomized list to package bottles of 5-day courses of either clindamycin or placebo, masked to the participants and investigators. The investigators allocated the bottles consecutively to the participants. The trial pharmacist retained the code for group allocation within a sealed envelope until all of the study data had been collected and analyzed. A total of 494 women were randomly assigned.⁷ The 126 participants in this subanalysis were all from 1 of the 2 participating centers, St. George’s Hospital. Women were excluded from random assignment if they had multiple pregnancy; needed or had cervical cerclage; history of cone biopsy; uterine, cervical, or fetal anomaly; diabetes, renal disease, lupus, antiphospholipid syndrome, or hypertension; allergy to clindamycin; were less than 16 years of age; or reported a fishy smelling vaginal discharge, in which case they received treatment and further genitourinary screening for sexually transmitted pathogens.⁷ The local research ethics committees at both centers approved the study.

The attending midwives were instructed on the technique of obtaining intramembranous microbiologic samples of the placenta before the study was started. After each delivery or miscarriage, the midwife, wearing a pair of clean (but not necessarily sterile) gloves, laid out the placenta on a clean flat surface and separated the chorion from the amnion by peeling them apart. Care was taken not to contaminate the exposed chorioamnionic surfaces, and microbiologic samples were obtained with sterile swabs 4–6 cm away from the site of membrane rupture. The placenta was then placed in 10% neutral buffered formalin and allowed to fix. All the placental specimens remained fixed until the end of the study, when they were examined by one of the investigators (I.J.), who was blinded to the treatment groups.

A careful macroscopic examination of the placenta and cord was performed. Blocks of the extraplacental membranes were taken to include all 4 quadrants of the gestational sac by preparing a series of membrane rolls in such a way that the margin of the site of rupture was in the center of each roll. At least 3 pieces of the umbilical cord were sampled by taking cross sections from different points along its length, and at least 3 full-thickness blocks of central parts of the villous tissue were examined. If the placental disc was too thick to allow a complete block of tissue incorporating both the maternal and fetal surfaces to be placed in 1 cassette, full-thickness blocks were divided into 2. As well as blocks from macroscopically normal villous tissue, any focal lesion present was also sampled. Care was taken when sampling the villous tissue to include both the central and peripheral parts of individual placental lobules. Tissue sections were cut at a thickness of 4 µm and stained with hematoxylin and eosin. They were scanned using a low power x 4 objective and then screened systematically with x 10 objective, using higher powers as and when necessary for cellular identification.

The presence of groups of polymorphonuclear leukocytes were noted separately in the amnion and chorion of the membranes, in the decidua, fetal surface of the placenta, the walls of fetal chorionic vessels, umbilical cord, and in the subchorionic fibrin layer forming the roof of the intervillous space immediately above the chorionic plate on the fetal surface. Small numbers of polymorphonuclear leukocytes, especially in the decidua and subchorionic fibrin, are extremely common in uncomplicated term pregnancies,⁸ and are widely believed to be a physiologic phenomenon of no significance.⁹ In view of this, scattered individual polymorphonuclear leukocytes were not regarded as significant and therefore not included for diagnosis, but 2 or more groups of 8 or more polymorphonuclear leukocytes were used to diagnose inflammation in the amnion, chorion, or decidua, or a single localized area with a dense polymorphonuclear cell infiltration. In the umbilical cord a diagnosis of funisitis was made if polymorphonuclear leukocytes were present within the Wharton’s jelly around any of the umbilical vessels with or without diffuse polymorphonuclear cell infiltration of the Wharton’s jelly. No significance was attached to focal pavementing of polymorphonuclear leukocytes within umbilical vessels in the absence of infiltration of the vessel wall, and similarly, polymorphonuclear leukocytes seen within the vessel wall but not extending into Wharton’s jelly were not included for diagnosis.

When assessing the amnion of the extraplacental membranes, polymorphonuclear leukocytes associated with cervical mucous or blood were not included for diagnosis. In the chorionic layer, cells showing nuclear pyknosis or features of apoptosis were distinguished from a true polymorph infiltrate. Scattered or small groups of mononuclear cells are a normal component of the decidua and in the absence of polymorphonuclear leukocytes were not included for the diagnosis of inflammation. The incidence of histologic chorioamnionitis was compared between the clindamycin and placebo groups, in any of the decidua, membrane, fetal, or subchorionic fibrin areas assessed.

Microbiologic cultures and laboratory methods were based on the Public Health Laboratory Service Standard Operating Procedures (available on www.phls.org.uk and www.hpa.org.uk). Placental swabs were inoculated onto the following media: blood agar incubated in air plus CO₂, cysteine-lysine-electrolyte-deficient agar and mannitol salt agar in air, and neomycin blood agar anaerobically. Cysteine-lysine-electrolyte-deficient and mannitol salt agar plates were incubated overnight; blood agar and neomycin-blood agar plates were incubated for 48 hours at 37°C. Beta-hemolytic streptococci were identified to species level by standard techniques. Other organisms were identified to genus level only. Antibiotic sensitivities of isolates were assessed by disc testing (Stokes method).

The a priori hypothesis of the study was that there would be a reduction in the percentage of histologic chorioamnionitis in the treatment group. A power calculation indicated that 60 women in each of the treatment and placebo arms would detect a difference of 24% in the percentage of women having histologic chorioamnionitis, decreasing from 50% in the placebo group¹ to 26% in the treatment group, with 80% power and a 5% significance level. The proportion of women with placental inflammation was compared between the clindamycin and placebo groups using the ² test, and pregnancy outcomes were compared between women with and without inflammation by Fisher exact test due to small numbers. Logistic regression was used to adjust the relationship between inflammation and treatment group for possible confounding factors and to adjust the relationship between inflammation and pregnancy outcome for treatment group (having first collapsed outcome into 2 categories: spontaneous preterm delivery or miscarriage compared with all other outcomes).

	RESULTS

TOP ABSTRACT PATIENTS AND METHODS RESULTS DISCUSSION REFERENCES

 
Of the 126 participants, placental analyses were available for 122, of whom 62 (51%) were in the clindamycin group and 60 (49%) in the placebo group. Two placental specimens received by the pathology laboratory were labeled incorrectly, whereas 2 other requests for pathologic examination were received with no accompanying placental specimens. The baseline characteristics for each treatment group are shown in Table 1. There were no statistically significant differences in placental inflammation between the clindamycin and placebo groups in any of the decidua, membrane, fetal, or subchorionic fibrin areas assessed (Table 2). Because inflammation tends to decrease with increasing gestational age, the comparison of inflammation between treatment groups was adjusted for gestational age; no significant differences emerged. Similarly, adjusting the association between inflammation and treatment for ethnic origin or history of miscarriage had no effect on the significance of the results. There were no significant differences in the outcomes of pregnancy between women with and without inflammation, either before or after adjustment for treatment group (Table 3). The number of subjects on whom specific organisms were isolated, for the clindamycin and placebo group respectively, was: anaerobes (1 and 3), coliforms (3 and 6), group B Streptococcus (1 and 1), and diphtheroids (2 and 1).

	DISCUSSION

TOP ABSTRACT PATIENTS AND METHODS RESULTS DISCUSSION REFERENCES

 
Our results showed no significant difference in the incidence of inflammation in the maternal or fetoplacental tissues of women with abnormal vaginal flora treated with oral clindamycin compared with those who received placebo. There was, however, a lower rate of extraplacental membrane inflammation in treated women (25% compared with 41%). Thus, at present the precise mechanism of action of early oral clindamycin in reducing miscarriage and preterm labor remains unclear.

Intrauterine, but localized extra-amniotic colonization or inflammation of the decidua by bacterial vaginosis organisms may activate the cascade of biochemical events leading to preterm delivery before invasion of the amniotic cavity and the fetus. Clindamycin therapy may exert beneficial effect at this stage by eradicating the microorganisms, although admittedly we have not found any direct evidence for this. Alternatively, the anti-inflammatory properties of clindamycin^10,11 may modify the host cytokine response such that the cascade of labor is not activated. This effect would of course not be readily evident in studies of morphologic features of inflammation in the placenta. Little is known about the transplacental transfer of clindamycin. After intravenous administration of clindamycin to the mother at term, peak levels in the cord blood are approximately 40% of peak maternal levels.¹² Clindamycin did not appear in detectable concentrations in the amniotic fluid during the first hour even though peak cord blood level was achieved within 20 minutes of maternal administration.¹² This is probably because clindamycin is metabolized mainly in the liver, and only a small amount is excreted by the kidney. Whether a greater transplacental transfer or excretion into the amniotic fluid through the fetal kidneys might enhance efficacy in treating chorioamnionitis is speculative. Furthermore, the participants in our study took oral clindamycin or placebo early in the second trimester (mean gestation 15.6 weeks), a time when fetal urine production and therefore renal excretion of clindamycin into the amniotic fluid is limited.

Yoon et al¹³ defined chorioamnionitis biochemically as amniotic fluid levels of IL-6 greater than 2.6 ng/mL. Raised amniotic fluid inflammatory cytokines predict adverse outcome better than positive amniotic fluid culture and clinical chorioamnionitis,^13,14 suggesting that fetal inflammatory response is more harmful and predictive of fetal injury than colonization or infection by microorganisms and maternal response. Umbilical vasculitis (funisitis) and angitis of the chorionic vessels represent histologic evidence of fetal inflammation and its response to infection. Oral clindamycin did not reduce the incidence of markers of fetal inflammation (funisitis and angitis) in treated women 16% (10/62) compared with placebo 14% (8/58). The relatively small size of the groups means that a real effect cannot be ruled out by this study, and in particular, a lower rate of membrane inflammation was observed in the clindamycin group compared with placebo (25% compared with 41%).

The groups were not completely balanced for ethnicity or history of previous miscarriage or preterm delivery. There were fewer women with a history of miscarriage or preterm delivery, and of Afro-Caribbean origin in the clindamycin group, which could have potentially led to a spurious association with inflammation, but because our results were not significant, this is not an issue.

Histologic chorioamnionitis is a postpartum diagnosis and has limited value as a basis for clinical intervention. Moreover, up to 36% of normal term placentas delivered by elective cesarean and 54% delivered vaginally, will display histologic evidence of chorioamnionitis.⁸ Other noninfective causes of chorioamnionitis, such as hypoxia, changes in amniotic fluid pH, maternal immunologic reactions, irritant effects of meconium, and nonspecific reactions to poor fetal condition, have been proposed.¹⁵

We conclude that although early oral clindamycin reduced the risk of late miscarriage and preterm birth in women with abnormal vaginal flora diagnosed in the second trimester of pregnancy, the results of the current analysis (albeit with small numbers in the groups) suggest that this effect is unlikely to be mediated through a reduction in the incidence of histologic chorioamnionitis.

	Footnotes

 
Funded by the NHS Executive London, Research and Development Programme. The authors thank Pharmacia-Upjohn for donation of the clindamycin and placebo used in this study and Dr. Aodhan Breathnach for his assistance in retrieving microbiologic data.

Presented at the 8th World Congress for Infectious and Immunological Diseases in Obstetrics and Gynaecology, Palazzo del Casino, Venice, Italy, November 8–11, 2003.

Corresponding author: Austin Ugwumadu, Department of Obstetrics and Gynecology, St. George’s Hospital, Blackshaw Road, London SW17 0QT, United Kingdom; e-mail: augwumad{at}sgul.ac.uk.

Financial Disclosure Phillip Hay has received payments for lectures and consultancy from Osmetech, 3M, and Pharmacia and Upjohn and has received funding for trials and to attend conferences from these companies. Austin Ugwumadu, Fiona Reid, Isaac Manyonda, and Iona Jeffrey have no disclosures to declare.

doi:10.1097/01.AOG.0000202399.13074.98

	REFERENCES

TOP ABSTRACT PATIENTS AND METHODS RESULTS DISCUSSION REFERENCES

 
1. Hillier SL, Martius J, Krohn M, Kiviat N, Holmes KK, Eschenbach DA. A case-control study of chorioamnionic infection and histologic chorioamnionitis in prematurity. N Engl J Med 1988;319:972–8.[Abstract]

2. Gomez R, Romero R, Ghezzi F, Yoon BH, Mazor M, Berry SM. The fetal inflammatory response syndrome. Am J Obstet Gynecol 1998;179:194–202.[Medline]

3. Williams MC, O’Brien WF, Nelson RN, Spellacy WN. Histologic chorioamnionitis is associated with fetal growth restriction in term and preterm infants. Am J Obstet Gynecol 2000;183:1094–9.[Medline]

4. Verma U, Tejani N, Klein S, Reale MR, Beneck D, Figueroa R, et al. Obstetric antecedents of intraventricular hemorrhage and periventricular leukomalacia in the low-birth-weight neonate. Am J Obstet Gynecol 1997;176:275–81.[Medline]

5. De Felice C, Toti P, Laurini RN, Stumpo M, Picciolini E, Todros T, et al. Early neonatal brain injury in histologic chorioamnionitis. J Pediatr 2001;138:101–4.[Medline]

6. Romero R, Sirtori M, Oyarzun E, Avila C, Mazor M, Callahan R, et al. Infection and labor. V. Prevalence, microbiology, and clinical significance of intraamniotic infection in women with preterm labor and intact membranes. Am J Obstet Gynecol 1989;161:817–24.[Medline]

7. Ugwumadu A, Manyonda I, Reid F, Hay P. Effect of early oral clindamycin on late miscarriage and preterm delivery in asymptomatic women with abnormal vaginal flora and bacterial vaginosis: a randomised controlled trial. Lancet 2003;361:983–8.[Medline]

8. Salafia CM, Weigl C, Silberman L. The prevalence and distribution of acute placental inflammation in uncomplicated term pregnancies. Obstet Gynecol 1989;73:383–9.[Abstract/Free Full Text]

9. Infections and inflammatory lesions of the placenta. In: Fox H. Pathology of the placenta. London (UK): W.B. Saunders; 1997. p. 294–343.

10. Esterly NB, Furey NL, Flanagan LE. The effect of antimicrobial agents on leukocyte chemotaxis. J Invest Dermatol 1978;70:51–5.[Medline]

11. Garraffo R, Jambou D, Lapalus P. Effects of antibiotics on the production of prostaglandins by macrophages in vitro [in French]. Pathol Biol (Paris) 1989;37:643–648.[Medline]

12. Antimicrobial agents. In: Sweet RL, Gibbs RS. Infectious diseases of the female genital tract. 4th ed. Philadelphia (PA): Lippincott Williams & Wilkins; 2002. p. 609-60.

13. Yoon BH, Romero R, Moon JB, Oh SY, Han SY, Kim JC, et al. The frequency and clinical significance of intra-amniotic inflammation in patients with a positive cervical fetal fibronectin. Am J Obstet Gynecol 2001;185:1137–42.[Medline]

14. Redline RW, Wilson-Costello D, Borawski E, Fanaroff AA, Hack M. The relationship between placental and other perinatal risk factors for neurologic impairment in very low birth weight children. Pediatr. Res 2000;47:721–26.[Medline]

15. Fox H. Pathology of the placenta. Philadelphia (PA): WB Saunders; 1978.

This Article Abstract Full Text (PDF) Alert me when this article is cited Alert me if a correction is posted Services Similar articles in this journal Similar articles in PubMed Alert me to new issues of the journal Download to citation manager Citing Articles Citing Articles via Google Scholar Google Scholar Articles by Ugwumadu, A. Articles by Jeffrey, I. Search for Related Content PubMed PubMed Citation Articles by Ugwumadu, A. Articles by Jeffrey, I. Related Collections General obstetrics Infectious disease Obstetric complications of pregnancy Pathology Placenta