Accuracy of the TDx-FLM assay of amniotic fluid: a comparison of vaginal pool samples with amniocentesis

Objective: The purpose of this study was to evaluate the use of the TDx-FLM fluorescence polarization assay on vaginal pool fluid in patients with preterm premature rupture of membranes (PPROM). Methods: A prospective matched-pairs study was performed at a tertiary care center. For each patient enrolled, amniotic fluid samples were obtained by sterile speculum examination and by amniocentesis within 12 h of each other. Inclusion criteria were the presence of PPROM and a gestational age of 30-36 weeks. The samples were analyzed separately using the TDx-FLM assay in the same laboratory. The results were compared using a paired Student t test. Results: A total of 16 patients received both amniocentesis and vaginal collection of amniotic fluid. The mean gestational age at amniocentesis was 33.3 weeks (SD 1.9). In every case, the vaginal pool TDx-FLM result was lower than the amniocentesis result. The mean difference in the assays between the two fluid sources was 35% (range 17-63%, p < 0.001). Amniocentesis suggested a mature result in 12 cases (75%), an indeterminate result in two cases (12.5%), and an immature result in two cases (12.5%). Vaginal pool fluid suggested a mature result in four cases (25%), an indeterminate result in nine cases (56%), and an immature result in three cases (19%). Using the cut-off values validated for amniocentesis specimens as a standard for comparison, vaginal pool TDx-FLM assay had 42% sensitivity, 100% specificity, 100% positive predictive value and 36% negative predictive value for predicting lung maturity. Conclusions: The TDx-FLM assay on vaginal pool samples of amniotic fluid yielded results that were significantly different from those of amniocentesis samples. At this point, the assay is only clinically useful for vaginal pool samples when a mature result is obtained.     

解脲支原体宫内感染与胎膜早破关系的研究

目的 探讨解脲支原体(UU)宫内感染与胎膜早破的相关性。方法 运用培养法对妊娠晚期32例胎膜早破孕妇(试验组)和20例正常孕妇(对照组)的剖宫产术中羊水进行支原体培养检测,评价6种抗菌药物对抑制支原体的敏感性;同时运用PCR-微板核酸分子杂交法对上述羊水进行解脲支原体DNA检测。结果 试验组分离培养检测的UU感染阳性率为28.1％,对照组UU感染阳性率为5.0％(P>0.05);6种抗生素药敏试验提示氧氟沙星药物敏感性最弱;在同时进行的PCR-微板核酸分子杂交法试验中,试验组的UU检出率为46.9％,对照组的UU检出率为10.0％,两组比较差异有显著性(P<0.01)。结论 解脲支原体所致的宫内感染是部分妊娠晚期胎膜早破的原因之一。PCR-微板核酸分子杂交法联合培养法,是诊断解脲支原体宫内感染的特异、快速和准确的方法。     

Correlation of quantitative amniotic fluid cultures with endometritis after cesarean section

At the time of cesarean section, amniotic fluid was collected transabdominally from 60 patients, and quantitative cultures were performed on the amniotic fluid. A culture was defined as positive if ≥10² colony-forming units per milliliter of a high-virulence organism were isolated. Any other result was defined as negative. In 24 patients with no labor or rupture of the membranes, no positive cultures were found, but there was a 25% incidence of endometritis. Among 36 patients with labor or rupture of the membranes, or both, 13 (36%) had a positive culture. Twelve of the 13 (92%) developed endometritis, whereas nine of the 23 (39%) patients with a negative culture had endometritis (p < 0.002). The usual clinical risk factors for endometritis were not different between the positive and negative culture groups. However, the patients with positive cultures had a significantly shorter time interval from cesarean section to endometritis than did the patients with negative cultures (p < 0.02). There was an excellent correlation between a positive amniotic fluid culture and endometritis after cesarean section.     

131例妊娠期肝内胆汁淤积症的围生结局分析

目的:探讨及早发现羊水胎粪污染及适时终止妊娠对妊娠期肝内胆汁淤积症(ICP)围生结局的影响.方法:对131例ICP病例资料进行回顾性分析.结果:共发现羊水胎粪污染49例,其中羊膜腔穿刺发现6例,人工破膜发现34例,剖宫产手术中发现4例,自然破膜发现5例.羊水污染率37.4%,早产率26.0%,剖宫产率62.6%,均高于同期分娩的非ICP患者(P0.05).3例新生儿Apgar 1分钟评分小于7分,无围生儿死亡.围生儿死亡率低于一些较大样本的文献累计的ICP围生儿死亡率(P0.05).结论:及早发现羊水胎粪污染及适时终止妊娠,对降低ICP患者围生儿死亡率具有重要的临床价值.     

Interleukin-6 and interleukin-1 receptor antagonist in amniotic fluid and cord blood in patients with pre-term, premature rupture of the membranes.

OBJECTIVE: We investigated the role of interleukin (IL)-6 and IL-1 receptor antagonist (IL-1ra) in pre-term premature rupture of the membranes (PROM). METHOD: Amniotic fluid samples were collected from 10 patients with pre-term PROM (group 1a), 13 patients undergoing genetic amniocentesis (group 2), seven patients with normal vaginal delivery (group 3a), and 11 patients with elective cesarean section (group 4a). Umbilical venous blood was collected from nine cases of pre-term PROM (group 1b), 19 cases of normal delivery (group 3b) and nine cases of elective cesarean section (group 4b). RESULTS: The concentration of IL-6 in the amniotic fluid in group 1a was significantly higher than in group 2 (P<0.001), and that in group 3a was significantly higher than in group 4a (P<0.001). The concentration of IL-1ra in amniotic fluid in group 1a was significantly higher than in group 2 (P<0.001). The concentrations of IL-6 in umbilical venous blood were not significant among the groups. The concentration of IL-1ra in umbilical venous blood in group 1b was significantly higher than in group 3b (P<0.05) and group 4b (P<0.05). CONCLUSIONS: High concentrations of amniotic fluid IL-6 present in pre-term PROM, are indicative of intrauterine inflammation, probably due to sub-clinical infection. The high concentrations of IL-1ra in amniotic fluid and umbilical venous blood suggest the possibility that the presence of this anti-inflammatory cytokine serves to prevent the development of overt inflammation.     

Fetal and maternal amniocentesis complications

AIM: The aim of this study was to estimate the incidence of complications due to the prenatal invasive diagnosis performed by amniocentesis at the Department of Prenatal Diagnosis of the Second University of Naples. METHODS: A total of 1.580 patients who underwent amniocentesis from January 2001 to December 2003, were submitted to a telephone interview concerning the complications that occurred after the invasive diagnosis. RESULTS: Only 1.416 patients out of the total patients interviewed answered correctly to the questionnaire. The complications that occurred in the first 24 h from amniocentesis (early complications) included light contractions and lipothymia respectively in 8.3% and 6.7% of cases; losses of amniotic fluid took place in 1.06% while bleedings were observed in 0.85% of cases. The incidence of abortions occurred in the week following amniocentesis was 0.78%. Preterm labor occurred in only 6% of the patients submitted to amniocentesis, spontaneous labor at term in 43% and cesarean section in 51%. The Apgar index at birth was normal in 95.7% of cases. In the remaining 4.3% pathologies like hypoglycemia or respiratory distress, which promptly regressed, were observed. Early complications and the incidence of abortions were significantly and independently associated with the double needle puncture but not with maternal age or the placenta specimen obtained during amniocentesis. CONCLUSIONS: A careful observation of the technique and the protocol, as well as a careful selection of patients are necessary presuppositions in order to further decrease the occurrence of complications due to amniocentesis.     

Effect of maternal hydration on amniotic fluid volume

OBJECTIVE: To estimate the effects of maternal intravenous hydration on amniotic fluid volume in normal pregnancies. METHODS: Women undergoing an amniocentesis for the evaluation of fetal lung maturity before an elective cesarean delivery were eligible to participate. An amniotic fluid index (AFI) was obtained before the amniocentesis, and at the time of the amniocentesis the amniotic fluid (AF) volume was determined by diazo-dye reaction with subsequent spectrophotometric analysis of AF samples. If the AF sample drawn for fetal maturity studies was mature, the patient was hydrated with 1000 mL of balanced salt solution 30 minutes before her cesarean delivery. Amniotic fluid volume was subsequently estimated after the hydration by a repeat AFI. Amniotic fluid volume was directly measured at cesarean delivery and compared with the dye-determined volume. The pre- and posthydration AFI were also compared. RESULTS: A total of 17 women participated in the study between January 2001 and June 2001. Statistically significant increases in the AF volume and AFI were found. The prehydration median AF volume was 450 mL (range 250-953), and the median increase in AF volume was 188 mL (95% confidence interval [CI] 60, 254 mL; P <.001). Median AFI was 8.6 (range 5.8-17.8) with a median change in AFI of 1.7 cm (95% CI 1.1, 3.0; P <.001). CONCLUSION: Maternal intravenous hydration appears to increase both the actual and ultrasound-estimated AF volumes in normal third-trimester pregnancies.     

False-positive amniotic fluid cytology in a parturient with active genital herpes infection at term.

A patient who developed genital herpes infection diagnosed at 37 weeks' gestation is presented. Amniocentesis performed at 38 weeks' gestation revealed herpetic changes in fetal cells. Subsequent amniotic fluid viral culture was negative. The infant was delivered by elective primary cesarean section at 39 weeks' gestation, demonstrated no evidence of herpetic infection, and was well at a 3-month follow-up examination. Positive cytologic changes seen in epithelial cells obtained by amniocentesis do not necessarily indicate intrauterine infection of the fetus. Cytologic changes alone are inadequate, and viral isolation by culture or by immunofluorescence should be required before prophylactic cesarean section is abandoned in the treatment of the parturient with active genital herpes infection near term.     

The presence of hepatitis B surface antigen and deoxyribonucleic acid in amniotic fluid and cord blood

OBJECTIVE: It is uncertain whether neonatal infection with hepatitis B, despite treatment after delivery with immunoglobulin and vaccine, is the result of prior in utero transmission of the virus or treatment failure. Furthermore, the potential risk of hepatitis B transmission from the mother to the fetus at the time a genetic amniocentesis is performed is also a concern. In an attempt to better elucidate these controversies, amniotic fluid and cord blood specimens obtained from pregnant women positive for hepatitis B surface antigen were analyzed for the presence of hepatitis B surface antigen and hepatitis B deoxyribonucleic acid. STUDY DESIGN: This study was a prospective longitudinal analysis that identified hepatitis B surface antigen-positive patients who presented for amniocentesis. Cord blood was obtained from these patients at the time of delivery. Cord blood was also obtained from a group of hepatitis B surface antigen-positive patients for whom no amniocentesis was performed. All samples were analyzed for the presence of hepatitis B surface antigen and hepatitis B deoxyribonucleic acid. RESULTS: A total of 121 hepatitis B surface antigen-positive pregnant women were identified. In the 72 pregnancies in which amniocentesis was not performed, 18% of the cord blood samples were positive for hepatitis B surface antigen and 4% were positive for hepatitis B deoxyribonucleic acid. Of 47 amniocentesis fluid samples, 32% were positive for hepatitis B surface antigen but all were negative for hepatitis B virus deoxyribonucleic acid. Of 30 cord blood samples from patients who underwent an amniocentesis, 27% were positive for hepatitis B surface antigen, but all were negative for hepatitis B virus deoxyribonucleic acid. CONCLUSIONS: This study found that hepatitis B viral deoxyribonucleic acid is rarely present in cord blood and was not identified in amniotic fluid obtained by amniocentesis. This finding suggests that in utero transmission of the virus is rare prior to the onset of labor. These data further confirm the reports in the current literature that the risk of hepatitis B transmission to the fetus during amniocentesis is low. Because hepatitis B surface antigen can exist as an isolated entity devoid of nuclear material, in some cases this protein may be able to traverse the placental and amniotic membrane barrier in a manner similar to other proteins, such as alpha-fetoprotein. Recommendations for genetic amniocentesis in women positive for hepatitis B surface antigen are discussed.     

Routine assessment of amniotic fluid alpha-Fetoprotein in early second-trimester amniocentesis is no longer justified

BACKGROUND: Open fetal neural tube defects are often followed by an increase in alpha-Fetoprotein concentration in amniotic fluid. For over 25 years there has been a routine to measure amniotic fluid alpha-Fetoprotein in conjunction with early genetic amniocentesis. The efficacy of such a screening test in a low-risk population has been questioned but never evaluated in a Swedish population. METHODS: Data were reviewed retrospectively from all consecutive early second-trimester genetic amniocenteses from two hospitals during the years 1993-2003. Indications for the genetic amniocenteses were maternal age > or = 35 years, maternal anxiety or a history of fetal aneuploidy. A questionnaire was sent to all obstetric clinics in Sweden regarding current common policy and experience of routine amniotic fluid alpha-Fetoprotein measurements, in the detection of open fetal neural tube defects. RESULTS: A total of 1,813 samples were included. In eight cases (0.4%) the amniotic fluid alpha-Fetoprotein concentrations were > or = 3 multiples of median, but five of them were false positive (63%). Out of the three true positive cases, one had clinical relevance. In the other two cases the detection of open fetal neural tube defects was of subordinate importance. In Sweden, during 2004, 91% of the obstetric clinics performed routine assessment of amniotic fluid alpha-Fetoprotein at second-trimester genetic amniocentesis, but only 9% regarded the analysis useful in clinical practice. CONCLUSIONS: According to our results, routine measurement of amniotic fluid alpha-Fetoprotein in early second-trimester genetic amniocentesis, to rule out a risk of open fetal neural tube defects, does not seem justified. The clinical usefulness seems to be limited.     

Intrapartum oligohydramnios does not predict adverse peripartum outcome among high-risk parturients

OBJECTIVE: Oligohydramnios can be defined by an amniotic fluid index <5th percentile for gestational age or an amniotic fluid index ≤5.0 cm regardless of gestational age. The purpose of this prospective study was to determine whether oligohydramnios by either definition predicts accurately, in a high-risk population, the risks for cesarean section for fetal distress, Apgar score <7 at 5 minutes, and neonatal acidosis.STUDY DESIGN: An amniotic fluid index was obtained in 490 consecutive parturients with medical or obstetric complications and a reliable gestational age. After each delivery, an umbilical arterial blood gas analysis was obtained. Both measures of amniotic fluid index were rated as screening tests with use of sensitivity, specificity, predictive values, and receiver-operator characteristic curves.RESULTS: The incidences of cesarean section for fetal distress and umbilical arterial pH <7.00 were 14% and 1.8%, respectively. The 70 neonates delivered by cesarean section for distress, compared with the 420 without, had a significantly higher incidence of pH <7.00 (8.5% vs 0.7%, p = 0.0004, relative risk 5.0, 95% confidence interval 2.9 to 8.4). Sensitivity and positive predictive values of an amniotic fluid index <5th percentile for gestational age to predict pH <7.00 were 0.8% and 22%, respectively, and for an amniotic fluid index ≤5.0 cm, 0.5% and 11%, respectively. Receiver-operator characteristic curves indicate that an amniotic fluid index between 0 and 20 cm cannot predict accurately which parturients will have cesarean sections for distress or be delivered of a newborn with a low Apgar score at 5 minutes or a pH <7.10.CONCLUSION: Both criteria for oligohydramnios are poor predictors of adverse outcome for high-risk intrapartum patients. (Am J Obstet Gynecol 1997;176:1130-8.)     

The effect of filtration of amniotic fluid on the growth of Chlamydia trachomatis and Escherichia coli

There is increasing concern about Chlamydia trachomatis infection during pregnancy, because of reports of increased maternal, fetal, and neonatal risks. Amniotic fluid is known to possess antibacterial activity and has recently been shown to inhibit formation of chlamydial inclusions in McCoy cell culture. To further characterize the anti-chlamydial factor, we investigated the effect of filtering the fluid (0.45 microns pores) prior to incubation. Amniotic fluid was obtained from 12 women at term gestation, either by amniocentesis, or at cesarean section, Chlamydial inclusion formation was studied in McCoy cell cultures, and Escherichia coli growth was studied by a plate-count method. Filtered amniotic fluid had significantly less inhibitory activity against chlamydial inclusion formation than nonfiltered fluid did. Both filtered and nonfiltered amniotic fluid were equally effective in inhibiting E. coli colony growth. These data suggest that the chlamydial inhibitor in amniotic fluid does not pass through 0.45 microns pores and is larger than the bacterial inhibitor that was reported to be a peptide of low molecular weight.     

Prenatal diagnosis of the Rhesus D fetal blood type on amniotic fluid in daily practice

Objective To retrospectively examine the diagnostic accuracy of prenatal RhD blood type genotyping on amniotic fluid, using a combination of two polymerase chain reaction (PCR) methods in daily practice. Methods Amniotic fluid was obtained from women undergoing amniocentesis. Two PCR protocols were carried out in two different laboratories. We obtained the postnatal serological RhD status. In cases with differing prenatal and postnatal test results, we investigated the possible error source by different methods. Sensitivity, specificity and the predictive values were calculated. Results Prenatal RhD genotyping was applied in 1,640 cases, of which the postnatal serologic RhD status was obtained in 927. No discordance between both PCR methods occurred. In nine out of 927 cases differing results between PCR and serologic status were encountered. The sensitivity was 99.5%, the specificity 98.6%, and both positive and negative predictive values 99.1%. Conclusion Prenatal diagnosis of the fetal RhD blood type with PCR from amniotic fluid is highly accurate in daily practice and associated with a minimal sensitivity of 99.5% and a minimal specificity of 98.6%.     

Risk of fetal chromosomal anomalies in patients with elevated maternal serum alpha-fetoprotein.

When elevated maternal serum alpha-fetoprotein (MSAFP) results lead to diagnostic amniocentesis, a decision of whether to karyotype fetal cells must be made. We examined our experience with MSAFP screening in 71,563 unselected pregnancies in which karyotyping was performed when amniocentesis was done because of MSAFP elevations. A total of 727 women (1.0%) underwent amniocentesis because of elevated MSAFP values and among this group, seven chromosomal anomalies (incidence one in 104) were detected. Of the 727 women, 658 (91%) had normal amniotic fluid AFP. In this group, there were six (one in 109) chromosomally abnormal fetuses: three with triploidy, two with 47,XXX, and one with 46,XX,1q-. Among the 69 pregnancies with elevated amniotic fluid AFP, one fetal chromosomal anomaly (trisomy 13) was diagnosed. The incidence of all chromosomal anomalies observed in women undergoing amniocentesis because of elevated MSAFP is comparable to that reported in women 36 years of age undergoing testing because of advanced maternal age. We believe that chromosome analysis should be performed on amniotic fluid samples obtained because of elevated MSAFP unless there are compelling financial circumstances that preclude this. Even in such cases, cell cultures should be established until the amniotic fluid AFP result is available. Chromosome analysis is essential when the amniotic fluid AFP is elevated because of the known association between open fetal defects (spina bifida, omphalocele, and scalp defects) and trisomies 13 and 18.     

False-positive acetylcholinesterase with early amniocentesis

Among 93 acetylcholinesterase determinations performed on amniotic fluid samples from pregnancies at 11-14 weeks' gestation, five unexplained false-positive results were observed. In four of the five cases, the ratio of acetylcholinesterase to pseudocholinesterase was compatible with that observed in association with open neural tube defects in later gestation. In contrast, no false-positive results were noted among 951 acetylcholinesterase determinations performed on samples from women at 15-20 weeks' gestation. Repeat amniocentesis was performed several weeks after the first procedure in four of the five cases of early amniocentesis and false-positive results; in each case, the acetylcholinesterase was negative on the second sample. All four pregnancies had a normal outcome. In the remaining case, trisomy 21 was diagnosed in the fetus and the pregnancy was terminated. Positive acetylcholinesterase results should be interpreted cautiously in samples from early amniocentesis, especially when the amniotic fluid alpha-fetoprotein level is not markedly elevated. The acetylcholinesterase-to-pseudocholinesterase ratio is not useful in identifying fetal neural tube defects before 15 weeks' gestation. Repeat amniocentesis may help in determining the significance of a positive acetylcholinesterase result from early amniocentesis when no fetal defect is identified by ultrasonography.     

Detection of fetal risk in postmaturity

Ninety-seven postmature pregnancies were monitored by amnioscopy or amniocentesis (to determine presence or absence of meconium), oxytocin challenge tests (OCT), 24-hour urinary oestriol estimations and fetal movement counts. The colour of the amniotic fluid and the result of the OCT predicted almost all cases of fetal distress in labour and infants with low Apgar scores. Oestriol estimations and fetal movement counts predicted fetal distress only when combined with other positive tests. Of 50 patients with no abnormal test results, 49 had uneventful labours. The Caesarean section rate was not above average and all babies were liveborn.     

Concentration of free fatty acids (FFA) in amniotic fluid and maternal and cord serum in cases of intrauterine growth retardation

There were no efficacious methods of diagnosing the syndrome of intrauterine growth retardation (IUGR) before birth. Only about 50% of cases with this syndrome can be recognized by clinical examinations. The aim of our investigations was an attempt to determine free fatty acid level (FFA) in amniotic fluid as a biochemical test in prenatal diagnosis. The investigations were carried out in 49 pregnant women and their newborns in whom IUGR was diagnosed clinically. The control group consisted of 30 pregnants who had an uneventful pregnancy and normal labor. Samples of amniotic fluid were taken in the first stage of parturition (dilatation of cervical os +/- 4-5 cm) by vaginal amniocentesis or amniotomy during cesarean section. The concentration of FFA in samples of amniotic fluid and serum were determined by Ducombe's method. It was found that during pregnancy complicated by IUGR, the free fatty acid concentration in amniotic fluid was almost three times higher than in normal pregnancy (0.355 mmol/l and 0.125 mmol/l) respectively. These differences are statistically significant (t = 12.58 p less than 0.001). Thus, determination of FFA can be used as a biochemical test in prenatal diagnosis of IUGR.     

应用羊水置换术治疗产时羊水Ⅱ、Ⅲ度胎粪污染

目的 :研究产时应用羊水置换技术对治疗羊水Ⅱ、Ⅲ度胎粪污染的疗效。方法 :选择产时羊水Ⅱ、Ⅲ度胎粪污染的足月单胎头位分娩孕妇 88例 ,其中的 4 4例进行羊水置换术的作为研究组 (A组 ) ;未进行羊水置换操作的另外 4 4例作为对照组 (B组 ) ;另选羊水无胎粪污染的 4 4例孕妇作为实验研究的空白对照组 (C组 )。观察其VD波缓解率、剖宫产率、新生儿血气分析及Apgar评分。结果 :A组变异减速 (VD)波的缓解率高于B组 (6 4.0 0 % ,2 9.17% ,P <0 .0 5 ) ;A组新生儿出生后 1分钟的脐血血气的pH值较B组高 (P <0 .0 5 ) ;A组的剖宫产率 (45 .4 5 % )低于B组(6 8.18% )。结论 :羊水置换对产时羊水Ⅱ、Ⅲ度胎粪污染有显著的疗效 ,可以明显改善新生儿的酸中毒与预后 ,且可以降低剖宫产率。     

The relationship between acute inflammatory lesions of the preterm placenta and amniotic fluid microbiology.

OBJECTIVE: Our objective was to determine the relationship between microbial invasion of the amniotic cavity and the presence and severity of acute inflammatory lesions in the placenta. STUDY DESIGN: Placental histologic and amniotic fluid microbiologic studies were performed in 92 consecutive patients who were admitted with preterm labor and intact membranes and delivered within 48 hours after amniocentesis. RESULTS: The prevalence of a positive amniotic fluid culture was 38% (35 of 92). There was a strong association between the presence and severity of inflammation in the amnion, chorion-decidua, umbilical cord, and chorionic plate and the results of the amniotic fluid culture (p less than 0.0001 for each tissue section). Three patterns of inflammation of the chorion-decidua were identified: marginating, nonmarginating, and a mixed pattern. The marginating and the mixed patterns of inflammation were strongly associated with the presence of a positive amniotic fluid culture. Acute inflammation of the chorionic plate was the most sensitive indicator of a microbial invasion of the amniotic cavity (sensitivity 96.6%), and funisitis and umbilical vasculitis had the highest specificity (85.7%). CONCLUSION: The presence of acute inflammatory lesions of the chorioamniotic membranes can serve as a marker of microbial invasion of the amniotic cavity.     

A dipstick test for infection in preterm premature rupture of the membranes

Microbiological analysis of amniotic fluid obtained by amniocentesis is increasingly employed in preterm premature rupture of the membranes (PPROM) to predict those patients at increased risk of chorioamnionitis. Although liquor culture appears the most accurate predictor, a role exists for rapid but slightly less accurate testing. In this study, a dipstick (Cytur-test) which predicts the presence of leukocytes was evaluated in amniotic fluid. Twenty-one samples were obtained by amniocentesis from 25 women with PPROM at gestation 26-34 weeks. The dipstick predicted 5 out of 6 specimens with a positive culture and 12 of 14 with a negative culture (figure 1). Its accuracy as a predictor of a positive culture was similar to microscopy for polymorphs, with sensitivities of 83.3% and 80% respectively and specificities of 85.7% and 92.3%. Neither test proved as predictive of histological chorioamnionitis due to 6 patients delivering greater than 48 hours after amniocentesis who may have developed infection subsequent to the demonstration of sterile liquor.