OBSTETRICS
Histological Chorioamnionitis Associated with Preterm Prelabour Rupture of Membranes at Kingston General Hospital: A Practice Audit Bryden Magee, MD, Graeme Smith, MD, PhD, FRCSC Department of Obstetrics and Gynaecology, Queen’s University, Kingston ON
Abstract
Résumé
Objective: To determine the prevalence of histological chorioamnionitis associated with preterm prelabour rupture of membranes (PPROM) in women following spontaneous onset of labour, urgent delivery or planned delivery after 34 weeks’ gestation.
Objectif : Déterminer la prévalence de la chorioamnionite histologique associée à la rupture prématurée des membranes préterme (RPMP) chez les femmes à la suite de l’apparition spontanée du travail, d’un accouchement d’urgence ou d’un accouchement planifié après 34 semaines de gestation.
Methods: Charts of all women admitted to Kingston General Hospital with PPROM prior to 34 weeks’ gestation over five years were collected. Obstetrical outcomes and histopathology reports were reviewed.
Méthodes : Les dossiers de toutes les femmes admises, au cours d’une période de cinq ans, à l’hôpital général de Kingston en raison d’une RPMP avant 34 semaines de gestation ont été rassemblés. Les issues obstétricales et les rapports d’histopathologie ont fait l’objet d’une analyse.
Results: Two hundred forty-four women with PPROM were identified and reviewed. The majority of women (169; 69%) went into spontaneous labour and, of those, 24 (14%) had clinical chorioamnionitis and 79 (47%) had histological chorioamnionitis. Of the 45 women (18%) who required urgent delivery, 27 (60%) had clinical chorioamnionitis and 31 (69%) had histological chorioamnionitis. Only 26 of the original 244 women with PPROM (11%) were managed expectantly until 34 weeks’ gestation and then had a planned delivery. The prevalence of histological chorioamnionitis in this group whose placentas were sent for histopathologic review was 24%. Overall, the clinical suspicion of chorioamnionitis was found to be specific (91%) but not sensitive (37%) for identifying chorioamnionitis on the basis of histopathology. Conclusion: Histological chorioamnionitis complicates almost one half of all cases of PPROM that occur prior to 34 weeks’ gestation. Most women will progress to spontaneous labour or require urgent delivery for clinical chorioamnionitis or other complications related to ruptured membranes before reaching 34 weeks’ gestation. Only a subset of women remain pregnant long enough to have labour induced, but among those the prevalence of histological chorioamnionitis is lower (24%).
Key Words: Fetal membranes, premature rupture, chorioamnionitis, premature birth Competing Interests: None declared. Received on June 22, 2013 Accepted on July 15, 2013
Résultats : Deux cent quarante-quatre femmes présentant une RPMP ont été identifiées et leurs dossiers ont fait l’objet d’une analyse. La majorité des femmes (169; 69 %) ont connu un travail spontané et, de celles-ci, 24 (14 %) ont présenté une chorioamnionite clinique et 79 (47 %) ont présenté une chorioamnionite histologique. Chez les 45 femmes (18 %) qui ont nécessité un accouchement d’urgence, 27 (60 %) ont présenté une chorioamnionite clinique et 31 (69 %) ont présenté une chorioamnionite histologique. Seulement 26 des 244 femmes présentant une RPMP qui ont été identifiées à l’origine (11 %) ont fait l’objet d’une prise en charge non interventionniste jusqu’à 34 semaines de gestation, pour ensuite connaître un accouchement planifié. Au sein de ce groupe, la prévalence de la chorioamnionite histologique (dans les cas où le placenta a fait l’objet d’une analyse histopathologique) a été de 24 %. De façon globale, nous avons constaté que les soupçons cliniques à l’égard de la présence d’une chorioamnionite étaient spécifiques (91 %), mais non sensibles (37 %), pour ce qui est de l’identification de la chorioamnionite en fonction de l’histopathologie. Conclusion : La présence d’une chorioamnionite histologique complique près de la moitié de tous les cas de RPMP qui se manifestent avant 34 semaines de gestation. La plupart des femmes en viendront à connaître un travail spontané ou à nécessiter un accouchement d’urgence motivé par la présence d’une chorioamnionite clinique ou d’autres complications liées à la rupture des membranes avant 34 semaines de gestation. Seul un sous-ensemble de femmes demeurent enceintes assez longtemps pour pouvoir faire l’objet d’un déclenchement du travail; toutefois, chez ces femmes, la prévalence de la chorioamnionite histologique est moindre (24 %). J Obstet Gynaecol Can 2013;35(12):1083–1089
DECEMBER JOGC DÉCEMBRE 2013 l 1083
Obstetrics
INTRODUCTION
P
reterm prelabour rupture of membranes (PPROM) occurs in 2% to 3% of pregnancies and can have profound implications for the pregnancy outcomes of both mother and fetus.1 The most significant complications of PPROM are prematurity and intrauterine infection, both of which carry considerable risk of short- and long-term sequelae for the newborn.2 The use of antibiotics and corticosteroids in the expectant management of women with PPROM has been shown to reduce the risk of major neonatal morbidity in preterm infants.3,4 In keeping with these findings, it is our centre’s practice to manage women with PPROM expectantly until 34 weeks’ gestation, when labour is induced. However, the optimal time for delivery between 32 and 37 weeks is unclear. There is great variability among regional protocols and no consensus among academic centres.1,5–12 Many women who are managed expectantly deliver earlier because of the spontaneous onset of labour or because they develop clinical evidence of intrauterine infection, known as clinical chorioamnionitis. Chorioamnionitis is defined as infection of the chorioamnion and is thought to be both a cause and a consequence of PPROM.10 A multicentre, prospective study in 2009 showed that when adjusted for gestational age, chorioamnionitis in preterm infants is strongly associated with an increased risk of early sepsis as well as severe intraventricular hemorrhage.13 Elevated levels of cytokines in the blood and brain in response to maternal infection are thought to cause neurotoxic damage to fetal white matter. This produces the so-called fetal inflammatory response syndrome, leading to intraventricular hemorrhage, periventricular leukomalacia, and cerebral palsy.14 For this reason, the treatment of chorioamnionitis includes early administration of parenteral antibiotics and expeditious delivery. Antenatal signs and symptoms used to diagnose clinical chorioamnionitis often appear late, by which point the fetus may have already mounted an inflammatory response. At our institution, a protocol has been established to submit all placentas from cases of PPROM for histopathologic review. Histological chorioamnionitis is defined as the presence of infiltrating polymorphonuclear leukocytes within the amniotic membranes, chorionic plate, and/or umbilical cord. Although the association is not as strong as with clinical chorioamnionitis, there remains a clinically significant association between histological chorioamnionitis and periventricular leukomalacia leading to cerebral palsy in the preterm neonatal population.15 If there is no evidence of preterm labour, chorioamnionitis, or fetal distress, women who have a diagnosis of PPROM are admitted at our centre for expectant management 1084 l DECEMBER JOGC DÉCEMBRE 2013
with prophylactic antibiotics and a single course of corticosteroids.16 From this point onwards, three potential obstetrical outcomes become manifest: Group 1: Onset of spontaneous labour and delivery prior to 34 weeks’ gestation; Group 2: Change in maternal or fetal status requiring urgent delivery; Group 3: Expectant management until planned delivery at 34 weeks’ gestation. The primary goal of this study was to determine the risk of histological chorioamnionitis in women with PPROM associated with these three obstetrical outcome groups: spontaneous onset of labour (Group 1), need for urgent delivery (Group 2), or planned delivery at 34 weeks (Group 3). The timing of delivery cannot be modified in Groups 1 and 2, as these women declare themselves early. However, the timing of delivery for women in Group 3 may be modified if the results suggest a high risk of subclinical (i.e., histological) chorioamnionitis, as seen on histopathologic review of the placentas postpartum. METHODS
Data were collected from charts of all women admitted to Kingston General Hospital between January 2005 and December 2009 with PPROM before 34+0 weeks. To retrieve these data, a broad search was undertaken in the Better Outcomes Registry Network perinatal database to capture all women who delivered at less than 35 weeks’ gestation for any reason. All cases were then reviewed using paper and/or electronic hospital charts to include only those who presented with PPROM at less than 34+0 weeks. Discharge records, interdisciplinary records, procedure notes, and pathology reports were reviewed. Gestational age when rupture of membranes was confirmed and gestational age at delivery were recorded to calculate the latency period. Labour type (spontaneous vs. induced), delivery type (vaginal vs. Caesarean section), and presence of clinical and/or histological chorioamnionitis were recorded. This study was approved by the Queen’s University Health Sciences and Affiliated Teaching Hospitals Research Ethics Board. RESULTS
Between January 2005 and December 2009, 244 women presented to Kingston General Hospital with PPROM at or before 34 weeks’ gestation and subsequently delivered. Overall, 115 (47%) women had evidence of histological
Histological Chorioamnionitis Associated with Preterm Prelabour Rupture of Membranes at Kingston General Hospital: A Practice Audit
Figure 1. Prevalence of histological chorioamnionitis by gestational age: gestational age at rupture of membranes
Number of patients
120 100 80 60 40 20 0
18−19+6
20−21+6
22−23+6
24−25+6
26−27+6
28−29+6
30−31+6
32−33+6
Gestational age No histological chorioamnionitis
Histological chorioamnionitis
Not reviewed by pathology
Figure 2. Prevalence of histological chorioamnionitis by gestational age: gestational age at delivery
Number of patients
120 100 80 60 40 20 0
18−19+6
20−21+6
22−23+6
24−25+6
26−27+6
28−29+6
30−31+6
32−34+
Gestational age No histological chorioamnionitis
Histological chorioamnionitis
chorioamnionitis. However, the prevalence of histological chorioamnionitis was highest among the women who presented with ruptured membranes and delivered at between 22 and 29+6 weeks’ gestation, as shown in Figures 1 and 2. Each woman had one of the three obstetrical outcomes as outlined in Figure 3. Group 1
The majority of women (169; 69%) went into labour spontaneously. As illustrated in Figure 4, 24 of these women were also found to have clinical chorioamnionitis while in labour. Histopathologic review of the placentas postpartum showed that 21 (88%) had evidence of histological chorioamnionitis and three (12%) did not. The remaining 145 (86%) women who went into spontaneous labour had no clinical signs or symptoms of chorioamnionitis. Within this group, 119 placentas were sent for histopathologic review, of which 58 (49%) had
Not reviewed by pathology
evidence of histological chorioamnionitis. Twenty-six placentas were not sent for pathological review. The prevalence of histological chorioamnionitis among women who laboured spontaneously after variable latency periods (time between rupture of membranes and spontaneous onset of labour leading to delivery) is shown in Figure 5. Over one third of these women delivered within one day of rupture of membranes and were found to have the lowest prevalence of chorioamnionitis (27%). Women who had a longer latency followed by spontaneous onset of labour were more likely to have histological chorioamnionitis. Group 2
Forty-five (18%) women required urgent delivery for a variety of indications, outlined in Figure 6. Twenty-seven (60%) were delivered urgently after clinical chorioamnionitis was diagnosed based on a combination DECEMBER JOGC DÉCEMBRE 2013 l 1085
Obstetrics
Figure 3. Outcomes of preterm prelabour rupture of membranes
PPROM < 34 weeks 244 Group 1 Spontaneous labour
24
No clinical chorio 145
30
45
169 Clinical chorio
Group 3 Planned delivery
Group 2 Urgent delivery
Clinical chorio 27
> 34 weeks
Other 18
Other
26
4
Figure 4. Prevalence of clinical and histological chorioamnionitis in Group 1
Number of patients
100% 80% 60% 40% 20% 0%
Clinical chorioamnionitis n = 24
No histological chorioamnionitis
Histological chorioamnionitis
of maternal signs and symptoms, leukocyte count, and fetal heart rate. Of these women, 21 (78%) were confirmed as having histological chorioamnionitis. Five did not have histological chorioamnionitis, and one placenta was not reviewed. Six women had urgent delivery because of abnormal fetal heart rate tracings, specifically recurrent decelerations and terminal bradycardia. Five women were delivered because of antepartum hemorrhage. Two women had umbilical cord prolapse, requiring urgent Caesarean section. Five women had labour induced for other reasons: herpes simplex virus infection, a foot presentation, refractory uterine irritability, decreasing insulin requirements (in an insulin-dependent diabetic), and failure to progress in labour. In total, 31 of the 45 women who required urgent delivery for any reason had histological chorioamnionitis. 1086 l DECEMBER JOGC DÉCEMBRE 2013
No clinical chorioamnionitis n = 145 Not reviewed by pathology
Group 3
Thirty women went on to have a planned delivery. Four women had labour induced before 34 weeks for reasons other than PPROM: three for IUGR, and one who requested medical termination at 23 weeks rather than expectant management. Therefore, only 26 (11%) of the original 244 women remained pregnant at 34 weeks’ gestation and were eligible for a planned delivery, according to our centre’s protocol. Only 17 placentas were sent for review. Of these, four (24%) had evidence of histological chorioamnionitis and 13 (76%) did not. As outlined in the Table, of all women presenting with PPROM at less than 34 weeks the clinical diagnosis of chorioamnionitis was made in 20%, while the histological diagnosis was made in 47%. Thus, clinical chorioamnionitis is specific (91%) but not sensitive (37%) for identifying cases of histological chorioamnionitis, as described in the Table.
Histological Chorioamnionitis Associated with Preterm Prelabour Rupture of Membranes at Kingston General Hospital: A Practice Audit
Figure 5. Latency period and prevalence of histological chorioamnionitis in Group 1 70
Number of patients
60 50 40 30 20 10 0
≤ 1 day
2 to 7 days
1 to 2 wks
2 to 3 wks
3 to 4 wks
> 4 wks
Latency period No histological chorioamnionitis
Histological chorioamnionitis
DISCUSSION
The results of this observational study expose several trends in the rates of histological chorioamnionitis. The risk of histological chorioamnionitis was found to be inversely related to gestational age at time of rupture of membranes and delivery, i.e., the earlier the gestational age, the higher the risk of histological chorioamnionitis. The rates of histological chorioamnionitis also varied markedly when stratified by obstetrical outcome, as seen in the Table. The prevalence of histological chorioamnionitis was high in Groups 1 and 2, as predicted on the basis of the pathophysiology of PPROM being both a cause and consequence of chorioamnionitis, which is a strong stimulus for labour. These results support clinical practice at our centre, where women with PPROM receive empiric antibiotic therapy but are not treated with tocolysis because the onset of labour may in fact be the body’s natural response to significant intra-amniotic infection. Besides preterm labour, almost 20% of women developed other complications of PPROM, all of which resulted in preterm delivery. Clinical chorioamnionitis was the most common, but other known complications occurred, including antepartum hemorrhage, abnormal fetal heart rate tracings, and cord prolapse. Our study found that 68% of women who were delivered urgently for any indication had histological chorioamnionitis, which contributes to the short- and long-term morbidity of the premature neonate. One important objective was to determine the prevalence of histological chorioamnionitis in women with PPROM electively delivered at 34 weeks, according to our centre’s current protocol. Interestingly, only 11% of women
Not reviewed by pathology
remained safely pregnant long enough to consider elective delivery at 34 weeks. In this small percentage of women with no other indications for delivery, the prevalence of histological chorioamnionitis was found to be 24%. To our knowledge, this is the first study that quantifies that risk, and our findings may aid physicians who struggle to balance the risks of chorioamnionitis and prematurity in cases of PPROM. It would be interesting to compare our rates of histological chorioamnionitis with those of another centre that has a different protocol (e.g., a centre in which women with PPROM are managed expectantly until 36 weeks). A national, multicentred study would be a useful next step. One of the limitations of our study is that 38 placentas (16%) were not sent for histopathologic review. The estimate of the risk of histological chorioamnionitis is especially limited in Group 3, because one third of placentas from this group were not examined. Clinicians may be falsely reassured by women with PPROM who have no evidence of clinical infection or labour. However, the clinical assessment for chorioamnionitis is not sensitive; that is, women without clinical signs or symptoms of chorioamnionitis are still at risk of intrauterine infection. This emphasizes the importance of sending all placentas for histopathologic review, because those that are infected often cannot be identified clinically. A further limitation of this study is that the diagnosis of clinical chorioamnionitis at our centre is not based on a standardized set of criteria, but is based instead on a combination of maternal signs and symptoms, leukocyte count, fetal heart rate changes, and clinical acumen. Women with clinical chorioamnionitis were identified for this study DECEMBER JOGC DÉCEMBRE 2013 l 1087
OBSTETRICS
Figure 6. Indications for urgent delivery 11%
Clinical chorioamnionitis
5%
Abnormal fetal heart rate 11% Antepartum hemorrhage
60%
Cord prolapse
13%
Other*
*Herpes simplex virus, foot presentation, decreasing insulin req ., refractory uterine irritabilty, failure to progress in labour
Prevalence of histological chorioamnionitis by group
Group 1: Spontaneous labour Group 2: Urgent delivery Group 3: Planned delivery
Clinical chorioamnionitis
Histological chorioamnionitis
Yes
21
3
0
No
58
61
26
Yes
21
5
1
No
10
6
2
Yes
–
–
–
No
5
16
9
by having written evidence of the diagnosis in their chart (progress notes or procedure suite records). The diagnosis was not assumed on the basis of biochemical values or vital signs. Some cases of clinical infection may therefore have been overlooked if the diagnosis was not written explicitly in the medical record. A final limitation is that we did not perform amniocentesis routinely for gram stain, culture, or other rapid biomarkers of infection. Although amniotic culture provides a definitive diagnosis of intra-amniotic infection, it requires an invasive procedure and can take up to 48 hours for results. The combination of rapid amniotic tests such as IL-6, gram stain, and leucocyte count lacks specificity, and therefore our centre has not found it to be superior to clinical diagnosis.17 CONCLUSION
Most pregnant women who presented with PPROM progressed to spontaneous labour or required urgent delivery for chorioamnionitis or other complications of ruptured membranes before reaching 34 weeks. Within these two groups, the risk of histological chorioamnionitis was high (51%). In the subset of women with PPROM 1088 l DECEMBER JOGC DÉCEMBRE 2013
No histological chorioamnionitis
Placenta not sent
who remained pregnant long enough to be considered for elective delivery, the prevalence of histological chorioamnionitis was lower (24%) at 34 weeks. However, this rate should be interpreted cautiously, because one third of the placentas in this group were not submitted for pathologic review. Histopathologic review of all placentas in cases of PPROM should be encouraged, regardless of the indication for delivery, because histological chorioamnionitis is prevalent and a known independent risk factor of neonatal morbidity. This information will better our understanding of the risks of expectant management versus urgent delivery and may contribute to a consensus on the optimal timing of delivery in women with PPROM. REFERENCES 1. Smith GN, Rafuse C, Anand N, Brennan B, Connors G, Crane J, et al. Prevalence, management, and outcomes of preterm prelabour rupture of the membranes of women in Canada. J Obstet Gynaecol Can 2005;27(6):547–53. 2. Patrick LA, Smith GN. Proinflammatory cytokines: a link between chorioamnionitis and fetal brain injury. J Obstet Gynaecol Can 2002;24(9):705–9. 3. Kenyon S, Boulvain M, Neilson JP. Antibiotics for preterm rupture of membranes. Cochrane Database Syst Rev 2010(8):001058.
Histological Chorioamnionitis Associated with Preterm Prelabour Rupture of Membranes at Kingston General Hospital: A Practice Audit
4. Roberts D, Dalziel S. Antenatal corticosteroids for accelerating fetal lung maturation for women at risk of preterm birth. Cochrane Database Syst Rev 2006;3:004454. 5. Buchanan SL, Crowther CA, Levett KM, Middleton P, Morris J. Planned early birth versus expectant management for women with preterm prelabour rupture of membranes prior to 37 weeks’ gestation for improving pregnancy outcome. Cochrane Database Syst Rev 2010(3):004735. 6. Mercer BM, Crocker LG, Boe NM, Sibai BM. Induction versus expectant management in premature rupture of the membranes with mature amniotic-fluid at 32 to 36 weeks—a randomized trial. Am J Obstet Gynecol 1993;169(4):775–82. 7. Hartling L, Chari R, Friesen C, Vandermeer B, Lacaze-Masmonteil T. A systematic review of intentional delivery in women with preterm prelabor rupture of membranes. J Matern Fetal Neonatal Med 2006;19(3):177–87. 8. Naef RW3rd, Allbert JR, Ross EL, Weber BM, Martin RW, Morrison JC. Premature rupture of membranes at 34 to 37 weeks’ gestation: aggressive versus conservative management. Am J Obstet Gynecol 1998;178(1 Pt 1):126–30. 9. Cox SM, Leveno KJ. Intentional delivery versus expectant management with preterm ruptured membranes at 30–34 weeks’ gestation. Obstet Gynecol 1995;86(6):875–9. 10. Garite TJ, Freeman RK. Chorioamnionitis in the preterm gestation. Obstet Gynecol 1982;59(5):539–45.
11. van der Ham DP, Nijhuis JG, Mol BW, van Beek JJ, Bijlenga D, Groenewout M, et al. Induction of labour versus expectant management in women with preterm prelabour rupture of membranes between 34 and 37 weeks (the PPROMEXIL-trial). BMC Pregnancy Childbirth 2007;7:11. doi:10.1186/1471–2393–7–11. 12. Al-Mandeel H, Alhindi M, Sauve R. Effects of intentional delivery on maternal and neonatal outcome in pregnancies with preterm prelabour rupture of membranes between 28 and 34 weeks of gestation: a systemic review and meta-analysis. J Matern Fetal Neonatal Med 2013;26(1):83–9. 13. Soraisham AS, Singhal N, McMillan DD, Sauve RS, Lee SK, Canadian Neonatal Network. A multicenter study on the clinical outcome of chorioamnionitis in preterm infants. Am J Obstet Gynecol 2009;200(4):372.e1,372.e6. 14. Wu YW, Escobar GJ, Grether JK, Croen LA, Greene JD, Newman TB. Chorioamnionitis and cerebral palsy in term and near-term infants. JAMA 2003; 290(20):2677–84. 15. Shatrov JG, Birch SC, Lam LT, Quinlivan JA, McIntyre S, Mendz GL. Chorioamnionitis and cerebral palsy: a meta-analysis. Obstet Gynecol 2010;116(2 Pt 1):387–92. 16. Society of Obstetricians and Gynaecologists of Canada. Antibiotic Therapy in Preterm Premature Rupture of the Membranes. SOGC Clinical Practice Guideline No. 233. J Obstet Gynaecol Can 2009;31(9):863–7. 17. Canavan TP, Simhan HN, Caritis S. An evidence-based approach to the evaluation and treatment of premature rupture of membranes: part I. Obstet Gynecol Surv 2004;59(9):669–77.
DECEMBER JOGC DÉCEMBRE 2013 l 1089
Histological Chorioamnionitis Associated with Preterm Prelabour Rupture of Membranes at Kingston General Hospital: A Practice Audit Bryden Magee, MD, Graeme Smith, MD, PhD, FRCSC Department of Obstetrics and Gynaecology, Queen’s University, Kingston ON
Abstract
Résumé
Objective: To determine the prevalence of histological chorioamnionitis associated with preterm prelabour rupture of membranes (PPROM) in women following spontaneous onset of labour, urgent delivery or planned delivery after 34 weeks’ gestation.
Objectif : Déterminer la prévalence de la chorioamnionite histologique associée à la rupture prématurée des membranes préterme (RPMP) chez les femmes à la suite de l’apparition spontanée du travail, d’un accouchement d’urgence ou d’un accouchement planifié après 34 semaines de gestation.
Methods: Charts of all women admitted to Kingston General Hospital with PPROM prior to 34 weeks’ gestation over five years were collected. Obstetrical outcomes and histopathology reports were reviewed.
Méthodes : Les dossiers de toutes les femmes admises, au cours d’une période de cinq ans, à l’hôpital général de Kingston en raison d’une RPMP avant 34 semaines de gestation ont été rassemblés. Les issues obstétricales et les rapports d’histopathologie ont fait l’objet d’une analyse.
Results: Two hundred forty-four women with PPROM were identified and reviewed. The majority of women (169; 69%) went into spontaneous labour and, of those, 24 (14%) had clinical chorioamnionitis and 79 (47%) had histological chorioamnionitis. Of the 45 women (18%) who required urgent delivery, 27 (60%) had clinical chorioamnionitis and 31 (69%) had histological chorioamnionitis. Only 26 of the original 244 women with PPROM (11%) were managed expectantly until 34 weeks’ gestation and then had a planned delivery. The prevalence of histological chorioamnionitis in this group whose placentas were sent for histopathologic review was 24%. Overall, the clinical suspicion of chorioamnionitis was found to be specific (91%) but not sensitive (37%) for identifying chorioamnionitis on the basis of histopathology. Conclusion: Histological chorioamnionitis complicates almost one half of all cases of PPROM that occur prior to 34 weeks’ gestation. Most women will progress to spontaneous labour or require urgent delivery for clinical chorioamnionitis or other complications related to ruptured membranes before reaching 34 weeks’ gestation. Only a subset of women remain pregnant long enough to have labour induced, but among those the prevalence of histological chorioamnionitis is lower (24%).
Key Words: Fetal membranes, premature rupture, chorioamnionitis, premature birth Competing Interests: None declared. Received on June 22, 2013 Accepted on July 15, 2013
Résultats : Deux cent quarante-quatre femmes présentant une RPMP ont été identifiées et leurs dossiers ont fait l’objet d’une analyse. La majorité des femmes (169; 69 %) ont connu un travail spontané et, de celles-ci, 24 (14 %) ont présenté une chorioamnionite clinique et 79 (47 %) ont présenté une chorioamnionite histologique. Chez les 45 femmes (18 %) qui ont nécessité un accouchement d’urgence, 27 (60 %) ont présenté une chorioamnionite clinique et 31 (69 %) ont présenté une chorioamnionite histologique. Seulement 26 des 244 femmes présentant une RPMP qui ont été identifiées à l’origine (11 %) ont fait l’objet d’une prise en charge non interventionniste jusqu’à 34 semaines de gestation, pour ensuite connaître un accouchement planifié. Au sein de ce groupe, la prévalence de la chorioamnionite histologique (dans les cas où le placenta a fait l’objet d’une analyse histopathologique) a été de 24 %. De façon globale, nous avons constaté que les soupçons cliniques à l’égard de la présence d’une chorioamnionite étaient spécifiques (91 %), mais non sensibles (37 %), pour ce qui est de l’identification de la chorioamnionite en fonction de l’histopathologie. Conclusion : La présence d’une chorioamnionite histologique complique près de la moitié de tous les cas de RPMP qui se manifestent avant 34 semaines de gestation. La plupart des femmes en viendront à connaître un travail spontané ou à nécessiter un accouchement d’urgence motivé par la présence d’une chorioamnionite clinique ou d’autres complications liées à la rupture des membranes avant 34 semaines de gestation. Seul un sous-ensemble de femmes demeurent enceintes assez longtemps pour pouvoir faire l’objet d’un déclenchement du travail; toutefois, chez ces femmes, la prévalence de la chorioamnionite histologique est moindre (24 %). J Obstet Gynaecol Can 2013;35(12):1083–1089
DECEMBER JOGC DÉCEMBRE 2013 l 1083
Obstetrics
INTRODUCTION
P
reterm prelabour rupture of membranes (PPROM) occurs in 2% to 3% of pregnancies and can have profound implications for the pregnancy outcomes of both mother and fetus.1 The most significant complications of PPROM are prematurity and intrauterine infection, both of which carry considerable risk of short- and long-term sequelae for the newborn.2 The use of antibiotics and corticosteroids in the expectant management of women with PPROM has been shown to reduce the risk of major neonatal morbidity in preterm infants.3,4 In keeping with these findings, it is our centre’s practice to manage women with PPROM expectantly until 34 weeks’ gestation, when labour is induced. However, the optimal time for delivery between 32 and 37 weeks is unclear. There is great variability among regional protocols and no consensus among academic centres.1,5–12 Many women who are managed expectantly deliver earlier because of the spontaneous onset of labour or because they develop clinical evidence of intrauterine infection, known as clinical chorioamnionitis. Chorioamnionitis is defined as infection of the chorioamnion and is thought to be both a cause and a consequence of PPROM.10 A multicentre, prospective study in 2009 showed that when adjusted for gestational age, chorioamnionitis in preterm infants is strongly associated with an increased risk of early sepsis as well as severe intraventricular hemorrhage.13 Elevated levels of cytokines in the blood and brain in response to maternal infection are thought to cause neurotoxic damage to fetal white matter. This produces the so-called fetal inflammatory response syndrome, leading to intraventricular hemorrhage, periventricular leukomalacia, and cerebral palsy.14 For this reason, the treatment of chorioamnionitis includes early administration of parenteral antibiotics and expeditious delivery. Antenatal signs and symptoms used to diagnose clinical chorioamnionitis often appear late, by which point the fetus may have already mounted an inflammatory response. At our institution, a protocol has been established to submit all placentas from cases of PPROM for histopathologic review. Histological chorioamnionitis is defined as the presence of infiltrating polymorphonuclear leukocytes within the amniotic membranes, chorionic plate, and/or umbilical cord. Although the association is not as strong as with clinical chorioamnionitis, there remains a clinically significant association between histological chorioamnionitis and periventricular leukomalacia leading to cerebral palsy in the preterm neonatal population.15 If there is no evidence of preterm labour, chorioamnionitis, or fetal distress, women who have a diagnosis of PPROM are admitted at our centre for expectant management 1084 l DECEMBER JOGC DÉCEMBRE 2013
with prophylactic antibiotics and a single course of corticosteroids.16 From this point onwards, three potential obstetrical outcomes become manifest: Group 1: Onset of spontaneous labour and delivery prior to 34 weeks’ gestation; Group 2: Change in maternal or fetal status requiring urgent delivery; Group 3: Expectant management until planned delivery at 34 weeks’ gestation. The primary goal of this study was to determine the risk of histological chorioamnionitis in women with PPROM associated with these three obstetrical outcome groups: spontaneous onset of labour (Group 1), need for urgent delivery (Group 2), or planned delivery at 34 weeks (Group 3). The timing of delivery cannot be modified in Groups 1 and 2, as these women declare themselves early. However, the timing of delivery for women in Group 3 may be modified if the results suggest a high risk of subclinical (i.e., histological) chorioamnionitis, as seen on histopathologic review of the placentas postpartum. METHODS
Data were collected from charts of all women admitted to Kingston General Hospital between January 2005 and December 2009 with PPROM before 34+0 weeks. To retrieve these data, a broad search was undertaken in the Better Outcomes Registry Network perinatal database to capture all women who delivered at less than 35 weeks’ gestation for any reason. All cases were then reviewed using paper and/or electronic hospital charts to include only those who presented with PPROM at less than 34+0 weeks. Discharge records, interdisciplinary records, procedure notes, and pathology reports were reviewed. Gestational age when rupture of membranes was confirmed and gestational age at delivery were recorded to calculate the latency period. Labour type (spontaneous vs. induced), delivery type (vaginal vs. Caesarean section), and presence of clinical and/or histological chorioamnionitis were recorded. This study was approved by the Queen’s University Health Sciences and Affiliated Teaching Hospitals Research Ethics Board. RESULTS
Between January 2005 and December 2009, 244 women presented to Kingston General Hospital with PPROM at or before 34 weeks’ gestation and subsequently delivered. Overall, 115 (47%) women had evidence of histological
Histological Chorioamnionitis Associated with Preterm Prelabour Rupture of Membranes at Kingston General Hospital: A Practice Audit
Figure 1. Prevalence of histological chorioamnionitis by gestational age: gestational age at rupture of membranes
Number of patients
120 100 80 60 40 20 0
18−19+6
20−21+6
22−23+6
24−25+6
26−27+6
28−29+6
30−31+6
32−33+6
Gestational age No histological chorioamnionitis
Histological chorioamnionitis
Not reviewed by pathology
Figure 2. Prevalence of histological chorioamnionitis by gestational age: gestational age at delivery
Number of patients
120 100 80 60 40 20 0
18−19+6
20−21+6
22−23+6
24−25+6
26−27+6
28−29+6
30−31+6
32−34+
Gestational age No histological chorioamnionitis
Histological chorioamnionitis
chorioamnionitis. However, the prevalence of histological chorioamnionitis was highest among the women who presented with ruptured membranes and delivered at between 22 and 29+6 weeks’ gestation, as shown in Figures 1 and 2. Each woman had one of the three obstetrical outcomes as outlined in Figure 3. Group 1
The majority of women (169; 69%) went into labour spontaneously. As illustrated in Figure 4, 24 of these women were also found to have clinical chorioamnionitis while in labour. Histopathologic review of the placentas postpartum showed that 21 (88%) had evidence of histological chorioamnionitis and three (12%) did not. The remaining 145 (86%) women who went into spontaneous labour had no clinical signs or symptoms of chorioamnionitis. Within this group, 119 placentas were sent for histopathologic review, of which 58 (49%) had
Not reviewed by pathology
evidence of histological chorioamnionitis. Twenty-six placentas were not sent for pathological review. The prevalence of histological chorioamnionitis among women who laboured spontaneously after variable latency periods (time between rupture of membranes and spontaneous onset of labour leading to delivery) is shown in Figure 5. Over one third of these women delivered within one day of rupture of membranes and were found to have the lowest prevalence of chorioamnionitis (27%). Women who had a longer latency followed by spontaneous onset of labour were more likely to have histological chorioamnionitis. Group 2
Forty-five (18%) women required urgent delivery for a variety of indications, outlined in Figure 6. Twenty-seven (60%) were delivered urgently after clinical chorioamnionitis was diagnosed based on a combination DECEMBER JOGC DÉCEMBRE 2013 l 1085
Obstetrics
Figure 3. Outcomes of preterm prelabour rupture of membranes
PPROM < 34 weeks 244 Group 1 Spontaneous labour
24
No clinical chorio 145
30
45
169 Clinical chorio
Group 3 Planned delivery
Group 2 Urgent delivery
Clinical chorio 27
> 34 weeks
Other 18
Other
26
4
Figure 4. Prevalence of clinical and histological chorioamnionitis in Group 1
Number of patients
100% 80% 60% 40% 20% 0%
Clinical chorioamnionitis n = 24
No histological chorioamnionitis
Histological chorioamnionitis
of maternal signs and symptoms, leukocyte count, and fetal heart rate. Of these women, 21 (78%) were confirmed as having histological chorioamnionitis. Five did not have histological chorioamnionitis, and one placenta was not reviewed. Six women had urgent delivery because of abnormal fetal heart rate tracings, specifically recurrent decelerations and terminal bradycardia. Five women were delivered because of antepartum hemorrhage. Two women had umbilical cord prolapse, requiring urgent Caesarean section. Five women had labour induced for other reasons: herpes simplex virus infection, a foot presentation, refractory uterine irritability, decreasing insulin requirements (in an insulin-dependent diabetic), and failure to progress in labour. In total, 31 of the 45 women who required urgent delivery for any reason had histological chorioamnionitis. 1086 l DECEMBER JOGC DÉCEMBRE 2013
No clinical chorioamnionitis n = 145 Not reviewed by pathology
Group 3
Thirty women went on to have a planned delivery. Four women had labour induced before 34 weeks for reasons other than PPROM: three for IUGR, and one who requested medical termination at 23 weeks rather than expectant management. Therefore, only 26 (11%) of the original 244 women remained pregnant at 34 weeks’ gestation and were eligible for a planned delivery, according to our centre’s protocol. Only 17 placentas were sent for review. Of these, four (24%) had evidence of histological chorioamnionitis and 13 (76%) did not. As outlined in the Table, of all women presenting with PPROM at less than 34 weeks the clinical diagnosis of chorioamnionitis was made in 20%, while the histological diagnosis was made in 47%. Thus, clinical chorioamnionitis is specific (91%) but not sensitive (37%) for identifying cases of histological chorioamnionitis, as described in the Table.
Histological Chorioamnionitis Associated with Preterm Prelabour Rupture of Membranes at Kingston General Hospital: A Practice Audit
Figure 5. Latency period and prevalence of histological chorioamnionitis in Group 1 70
Number of patients
60 50 40 30 20 10 0
≤ 1 day
2 to 7 days
1 to 2 wks
2 to 3 wks
3 to 4 wks
> 4 wks
Latency period No histological chorioamnionitis
Histological chorioamnionitis
DISCUSSION
The results of this observational study expose several trends in the rates of histological chorioamnionitis. The risk of histological chorioamnionitis was found to be inversely related to gestational age at time of rupture of membranes and delivery, i.e., the earlier the gestational age, the higher the risk of histological chorioamnionitis. The rates of histological chorioamnionitis also varied markedly when stratified by obstetrical outcome, as seen in the Table. The prevalence of histological chorioamnionitis was high in Groups 1 and 2, as predicted on the basis of the pathophysiology of PPROM being both a cause and consequence of chorioamnionitis, which is a strong stimulus for labour. These results support clinical practice at our centre, where women with PPROM receive empiric antibiotic therapy but are not treated with tocolysis because the onset of labour may in fact be the body’s natural response to significant intra-amniotic infection. Besides preterm labour, almost 20% of women developed other complications of PPROM, all of which resulted in preterm delivery. Clinical chorioamnionitis was the most common, but other known complications occurred, including antepartum hemorrhage, abnormal fetal heart rate tracings, and cord prolapse. Our study found that 68% of women who were delivered urgently for any indication had histological chorioamnionitis, which contributes to the short- and long-term morbidity of the premature neonate. One important objective was to determine the prevalence of histological chorioamnionitis in women with PPROM electively delivered at 34 weeks, according to our centre’s current protocol. Interestingly, only 11% of women
Not reviewed by pathology
remained safely pregnant long enough to consider elective delivery at 34 weeks. In this small percentage of women with no other indications for delivery, the prevalence of histological chorioamnionitis was found to be 24%. To our knowledge, this is the first study that quantifies that risk, and our findings may aid physicians who struggle to balance the risks of chorioamnionitis and prematurity in cases of PPROM. It would be interesting to compare our rates of histological chorioamnionitis with those of another centre that has a different protocol (e.g., a centre in which women with PPROM are managed expectantly until 36 weeks). A national, multicentred study would be a useful next step. One of the limitations of our study is that 38 placentas (16%) were not sent for histopathologic review. The estimate of the risk of histological chorioamnionitis is especially limited in Group 3, because one third of placentas from this group were not examined. Clinicians may be falsely reassured by women with PPROM who have no evidence of clinical infection or labour. However, the clinical assessment for chorioamnionitis is not sensitive; that is, women without clinical signs or symptoms of chorioamnionitis are still at risk of intrauterine infection. This emphasizes the importance of sending all placentas for histopathologic review, because those that are infected often cannot be identified clinically. A further limitation of this study is that the diagnosis of clinical chorioamnionitis at our centre is not based on a standardized set of criteria, but is based instead on a combination of maternal signs and symptoms, leukocyte count, fetal heart rate changes, and clinical acumen. Women with clinical chorioamnionitis were identified for this study DECEMBER JOGC DÉCEMBRE 2013 l 1087
OBSTETRICS
Figure 6. Indications for urgent delivery 11%
Clinical chorioamnionitis
5%
Abnormal fetal heart rate 11% Antepartum hemorrhage
60%
Cord prolapse
13%
Other*
*Herpes simplex virus, foot presentation, decreasing insulin req ., refractory uterine irritabilty, failure to progress in labour
Prevalence of histological chorioamnionitis by group
Group 1: Spontaneous labour Group 2: Urgent delivery Group 3: Planned delivery
Clinical chorioamnionitis
Histological chorioamnionitis
Yes
21
3
0
No
58
61
26
Yes
21
5
1
No
10
6
2
Yes
–
–
–
No
5
16
9
by having written evidence of the diagnosis in their chart (progress notes or procedure suite records). The diagnosis was not assumed on the basis of biochemical values or vital signs. Some cases of clinical infection may therefore have been overlooked if the diagnosis was not written explicitly in the medical record. A final limitation is that we did not perform amniocentesis routinely for gram stain, culture, or other rapid biomarkers of infection. Although amniotic culture provides a definitive diagnosis of intra-amniotic infection, it requires an invasive procedure and can take up to 48 hours for results. The combination of rapid amniotic tests such as IL-6, gram stain, and leucocyte count lacks specificity, and therefore our centre has not found it to be superior to clinical diagnosis.17 CONCLUSION
Most pregnant women who presented with PPROM progressed to spontaneous labour or required urgent delivery for chorioamnionitis or other complications of ruptured membranes before reaching 34 weeks. Within these two groups, the risk of histological chorioamnionitis was high (51%). In the subset of women with PPROM 1088 l DECEMBER JOGC DÉCEMBRE 2013
No histological chorioamnionitis
Placenta not sent
who remained pregnant long enough to be considered for elective delivery, the prevalence of histological chorioamnionitis was lower (24%) at 34 weeks. However, this rate should be interpreted cautiously, because one third of the placentas in this group were not submitted for pathologic review. Histopathologic review of all placentas in cases of PPROM should be encouraged, regardless of the indication for delivery, because histological chorioamnionitis is prevalent and a known independent risk factor of neonatal morbidity. This information will better our understanding of the risks of expectant management versus urgent delivery and may contribute to a consensus on the optimal timing of delivery in women with PPROM. REFERENCES 1. Smith GN, Rafuse C, Anand N, Brennan B, Connors G, Crane J, et al. Prevalence, management, and outcomes of preterm prelabour rupture of the membranes of women in Canada. J Obstet Gynaecol Can 2005;27(6):547–53. 2. Patrick LA, Smith GN. Proinflammatory cytokines: a link between chorioamnionitis and fetal brain injury. J Obstet Gynaecol Can 2002;24(9):705–9. 3. Kenyon S, Boulvain M, Neilson JP. Antibiotics for preterm rupture of membranes. Cochrane Database Syst Rev 2010(8):001058.
Histological Chorioamnionitis Associated with Preterm Prelabour Rupture of Membranes at Kingston General Hospital: A Practice Audit
4. Roberts D, Dalziel S. Antenatal corticosteroids for accelerating fetal lung maturation for women at risk of preterm birth. Cochrane Database Syst Rev 2006;3:004454. 5. Buchanan SL, Crowther CA, Levett KM, Middleton P, Morris J. Planned early birth versus expectant management for women with preterm prelabour rupture of membranes prior to 37 weeks’ gestation for improving pregnancy outcome. Cochrane Database Syst Rev 2010(3):004735. 6. Mercer BM, Crocker LG, Boe NM, Sibai BM. Induction versus expectant management in premature rupture of the membranes with mature amniotic-fluid at 32 to 36 weeks—a randomized trial. Am J Obstet Gynecol 1993;169(4):775–82. 7. Hartling L, Chari R, Friesen C, Vandermeer B, Lacaze-Masmonteil T. A systematic review of intentional delivery in women with preterm prelabor rupture of membranes. J Matern Fetal Neonatal Med 2006;19(3):177–87. 8. Naef RW3rd, Allbert JR, Ross EL, Weber BM, Martin RW, Morrison JC. Premature rupture of membranes at 34 to 37 weeks’ gestation: aggressive versus conservative management. Am J Obstet Gynecol 1998;178(1 Pt 1):126–30. 9. Cox SM, Leveno KJ. Intentional delivery versus expectant management with preterm ruptured membranes at 30–34 weeks’ gestation. Obstet Gynecol 1995;86(6):875–9. 10. Garite TJ, Freeman RK. Chorioamnionitis in the preterm gestation. Obstet Gynecol 1982;59(5):539–45.
11. van der Ham DP, Nijhuis JG, Mol BW, van Beek JJ, Bijlenga D, Groenewout M, et al. Induction of labour versus expectant management in women with preterm prelabour rupture of membranes between 34 and 37 weeks (the PPROMEXIL-trial). BMC Pregnancy Childbirth 2007;7:11. doi:10.1186/1471–2393–7–11. 12. Al-Mandeel H, Alhindi M, Sauve R. Effects of intentional delivery on maternal and neonatal outcome in pregnancies with preterm prelabour rupture of membranes between 28 and 34 weeks of gestation: a systemic review and meta-analysis. J Matern Fetal Neonatal Med 2013;26(1):83–9. 13. Soraisham AS, Singhal N, McMillan DD, Sauve RS, Lee SK, Canadian Neonatal Network. A multicenter study on the clinical outcome of chorioamnionitis in preterm infants. Am J Obstet Gynecol 2009;200(4):372.e1,372.e6. 14. Wu YW, Escobar GJ, Grether JK, Croen LA, Greene JD, Newman TB. Chorioamnionitis and cerebral palsy in term and near-term infants. JAMA 2003; 290(20):2677–84. 15. Shatrov JG, Birch SC, Lam LT, Quinlivan JA, McIntyre S, Mendz GL. Chorioamnionitis and cerebral palsy: a meta-analysis. Obstet Gynecol 2010;116(2 Pt 1):387–92. 16. Society of Obstetricians and Gynaecologists of Canada. Antibiotic Therapy in Preterm Premature Rupture of the Membranes. SOGC Clinical Practice Guideline No. 233. J Obstet Gynaecol Can 2009;31(9):863–7. 17. Canavan TP, Simhan HN, Caritis S. An evidence-based approach to the evaluation and treatment of premature rupture of membranes: part I. Obstet Gynecol Surv 2004;59(9):669–77.
DECEMBER JOGC DÉCEMBRE 2013 l 1089
