Thorson, Heidi L. MD; Ramaeker, Devon M. MD; Emery, Stephen P. MD
From the University of Pittsburgh Medical Center, Pittsburgh, Pennsylvania.
Support for this project was provided by the Department of Obstetrics and Gynecology, University of Pittsburgh Medical Center.
Presented at as a poster at the annual American Institute for Ultrasound in Medicine Conference, March 24–27, 2010, San Diego, California.
Corresponding author: Heidi L. Thorson, Magee Womens Hospital, Department of OB/GYN, 300 Halket Street, Pittsburgh, PA 15213; e-mail: firstname.lastname@example.org.
Financial Disclosure The authors did not report any potential conflicts of interest.
OBJECTIVE: To examine the association between interval from previous ultrasonogram to diagnosis of twin–twin transfusion syndrome and stage at diagnosis.
METHODS: We reviewed all monochorionic twins undergoing evaluation for twin–twin transfusion syndrome in our ultrasound department from 2001 to 2008. Cases of twin–twin transfusion syndrome were categorized as early (Quintero stages I and II) or late (Quintero stages III–V). The gestational age at diagnosis and the interval from the last ultrasonogram were recorded. The interval was categorized as 14 days or less or greater than 14 days. Pregnancies diagnosed with twin–twin transfusion syndrome at the initial ultrasonogram were categorized as greater than 14 days. The univariable association between interval from last scan and twin–twin transfusion syndrome stage was determined by χ2. P<.05 was considered significant.
RESULTS: One-hundred eight monochorionic twin pregnancies were identified, and 42 met criteria for twin–twin transfusion syndrome: 24 with early stage and 18 with late stage. The median gestational age at diagnosis was 19 6/7 (15 3/7–36 3/7) weeks. Twin–twin transfusion syndrome was more likely to be diagnosed at a later stage with an ultrasound interval greater than 14 days (P=.004). Two cases (11.8%) of late twin–twin transfusion syndrome were diagnosed in pregnancies with an ultrasound interval of 14 days or less.
CONCLUSION: An interval of greater than 14 days from previous ultrasonogram was strongly associated with detection of higher-stage twin–twin transfusion syndrome. These data suggest that a maximum surveillance interval of 14 days for monochorionic twins may lead to earlier stage at diagnosis, earlier intervention, and better outcomes.
LEVEL OF EVIDENCE: III
Twin–twin transfusion syndrome complicates up to 15% of monochorionic-diamniotic twin pregnancies.1,2 Twin–twin transfusion syndrome is a major cause of morbidity and mortality for monochorionic-diamniotic twins. Left untreated, mortality and morbidity rates approach 80–100%.1,3 The syndrome is the result of unbalanced blood flow through placental vascular anastomoses between the twins.4 The severity of the disease is classically staged using the Quintero staging system.5
Little is known about the natural history of twin–twin transfusion syndrome, such as gestational age of peak incidence or the rate of progression of the disease process. This lack of knowledge has resulted in inconsistent recommendations for initiation and interval of ultrasound screening of monochorionic-diamniotic twins. A crucial prerequisite is accurately recognizing a twin gestation as monochorionic. The American Institute of Ultrasound in Medicine, the American College of Radiology, the Royal College of Obstetricians and Gynaecologists (RCOG), and the National French College in Obstetrics and Gynecology (CNGOF) stress the importance of identifying and documenting chorionicity in all multiple gestations.6–9 Although the RCOG and the CNGOF also have recommended screening monochorionic gestations twice monthly for evidence of twin–twin transfusion syndrome, the American College of Obstetricians and Gynecologists has not made any recommendations.10 Because evidence suggests that intervention at an earlier Quintero stage leads to better pregnancy outcomes, defining and implementing the optimal surveillance strategy is critically important.11 In this retrospective cohort study we sought to examine the association of the stage at diagnosis of twin–twin transfusion syndrome, gestational age, and the interval from the previous ultrasonogram to diagnosis.
MATERIALS AND METHODS
We queried the Magee-Womens Hospital ultrasound database from the period 2001–2008 for all ultrasonograms with twin–twin transfusion syndrome as the indication or diagnosis for the ultrasonogram. Pregnancies were excluded if one or more fetus was identified as having a major congenital anomaly, a chromosomal abnormality, or if the pregnancy was monoamniotic. Pregnancies were also excluded if the ultrasound findings were more consistent with a process other than twin–twin transfusion syndrome as agreed to by two investigators on independent review (eg, preterm premature rupture of membranes, discordant uteroplacental insufficiency). Monochorionicity in our ultrasound unit is diagnosed in the first trimester by counting gestational sac number, placenta number, membrane thickness, and “T” sign. Diagnosis in the second or third trimester is made by placenta number, membrane thickness, “T” sign, and concordant sex.
During the study period, the screening protocol for twin–twin transfusion syndrome consisted of an ultrasound examination evaluating amniotic fluid volume and the presence or absence of bladder fullness in both twins. A complete anatomic survey including chorionicity was performed during the initial ultrasound examination in our department. Biometric measurements were performed if it was more than 3 weeks from their previous biometric evaluation. Arterial and venous Doppler interrogation was performed if an abnormality was identified on the ultrasound consistent with either twin–twin transfusion syndrome or uteroplacental insufficiency. Doppler evaluation included assessment of the umbilical artery, ductus venous, and umbilical vein in both twins.
The diagnosis of twin–twin transfusion syndrome was made based on Quintero staging and classified by two of the investigators independently.5 Stage I was defined as evidence of hydramnios in the sac of the recipient twin and oligohydramnios in the sac of the donor twin in the absence of additional findings. Hydramnios in the sac of the recipient twin was defined as deepest vertical pocket more than 8 cm. Oligohydramnios in the sac of the donor twin was defined by deepest vertical pocket measuring less than 2 cm. No pregnancies underwent amnioreduction before classifying the pregnancy with the initial diagnosis of twin–twin transfusion syndrome. The absence of a recognizable bladder or non-filling of the bladder in the donor twin defined stage II twin–twin transfusion syndrome. The location of the bladder was confirmed with color flow Doppler examination of the umbilical arteries separating in the fetal pelvis. Stage III was defined as critically abnormal Doppler studies in either donor or recipient. Stage IV required evidence of fetal hydrops. Stage V was death of one or both twins. Pregnancies fulfilling the diagnosis of twin–twin transfusion syndrome were then categorized as early-stage twin–twin transfusion syndrome (Quintero stages I and II) or late stage twin–twin transfusion syndrome (Quintero stages III, IV, V).
The gestational age at diagnosis and the interval from the last normal ultrasonogram before diagnosis of twin–twin transfusion syndrome were recorded. This interval was categorized a priori as 14 days or less or greater than 14 days. If a pregnancy was diagnosed with twin–twin transfusion syndrome at the initial ultrasonogram, it was categorized as greater than 14 days. Statistical analysis was performed using STATA 11.0. The univariable association between interval from last scan and stage of twin–twin transfusion syndrome was determined by χ2. P<.05 was considered significant. Multivariable logistic regression was performed to assess gestational age at diagnosis as an effect modifier.
This retrospective study of monochorionic twin pregnancies was approved by the Institutional Review Board of the University of Pittsburgh Medical Center.
Of the 108 monochorionic-diamniotic pregnancies identified on initial query, 42 (38.9%) met the criteria for twin–twin transfusion syndrome. Sixty-one pregnancies did not have a diagnosis of twin–twin transfusion syndrome, two pregnancies were excluded for major congenital anomalies, one pregnancy had a diagnosis of trisomy 21, and two pregnancies were confirmed to be dichorionic. Clinical details of the pregnancies are presented in Table 1. Twenty-four pregnancies were diagnosed with early (Quintero stage I and II) twin–twin transfusion syndrome and 18 with late (Quintero stages III–V) twin–twin transfusion syndrome. Twenty-four pregnancies had a previous ultrasonogram to calculate an interval between ultrasonograms ranging from 3–63 days (mean 19 plus or minus 15.1 days). Eighteen (42%) of cases were diagnosed at their initial ultrasonogram and categorized as greater than 14 days. The median gestational age at diagnosis was 19 6/7 weeks and ranged between 15 3/7 and 36 3/7 weeks gestation. Two-thirds (28 of 42) of all cases of twin–twin transfusion syndrome in our series were diagnosed before 22 0/7 weeks gestation with the peak incidence (9 of 42 cases, or 21.42%) between 18 0/7 and 18 6/7 weeks of gestation. Half were diagnosed by 20 weeks of gestation (Fig. 1). Finally, 22 of 42 (53%) of the cases were diagnosed at the time of their anatomic survey. Twin–twin transfusion syndrome was more likely to be diagnosed at a later stage with an ultrasound interval greater than 14 days (P=.004) (Fig. 2). The breakdown of twin–twin transfusion syndrome cases are shown in Table 2. Only two cases (11.8%) of late twin–twin transfusion syndrome were diagnosed in pregnancies with an ultrasound interval of 14 days or less (Table 2).
A screening interval greater than 14 days was associated with a late Quintero stage at diagnosis with an odds ration (OR) of 9.45 (95% confidence interval [CI] 1.8–50.4). When controlled for gestational age at diagnosis, a longer screening interval was still associated with a later Quintero stage with an adjusted OR of 9.8 (95% CI 1.8–52.9).
Our study suggests that a shorter surveillance interval in monochorionic-diamniotic gestations results in an earlier stage of twin–twin transfusion syndrome at diagnosis. These data support the findings from Sueters et al that twin–twin transfusion syndrome is diagnosed at an earlier stage when ultrasound surveillance is every 2 weeks combined with symptoms of hydramnios.12 Earlier stage at diagnosis has been shown in previous studies to be an important prognostic factor in successful pregnancy outcome.11,13 It is also well documented that perinatal survival decreases with increasing twin–twin transfusion syndrome stage.5,14–16 For example, the rate of survival in the Eurofetus trial for at least one twin was higher in pregnancies diagnosed at stage I or II (53 of 73 [73%]) compared with those diagnosed at stages III or IV (38 of 69 [55%]).11
The median gestational age for diagnosis of twin–twin transfusion syndrome in our series was 19 6/7 weeks, with a range of 15 3/7–36 3/7 weeks and a peak incidence of 18 0/7–18 6/7 weeks. Although two-thirds of the cases were diagnosed between 15 and 22 weeks, the other third were diagnosed between 22 and 36 weeks. This suggests that screening should begin before 18 weeks in known monochorionic-diamniotic gestations and screening should continue throughout the pregnancy until delivery. Given our findings of twin–twin transfusion syndrome diagnosed in the early second trimester and late third trimester, we recommend initiation of screening by 16 weeks and continuing ultrasound surveillance every 2 weeks until delivery. This is consistent with the RCOG and the CNGOF, which recommend ultrasound examinations at least twice a month for monochorionic-diamniotic pregnancies.6,7 Approximately half (53%) of the cases in our series were diagnosed with twin–twin transfusion syndrome at the time of fetal anatomic survey. More than half (24 of 42) of the cases of twin–twin transfusion syndrome had a previous ultrasonogram documenting a monochorionic-diamniotic gestation, yet the interval between ultrasonograms varied markedly from 3 to 63 days. Given what we know about the increased incidence of complications of monochorionic gestations, it is important to not only definitively diagnose monochorionic twin gestations (ie, not simply as “twins”) but to institute an earlier and more intensive screening strategy than in diamniotic gestations.17,18 Earlier detection of twin–twin transfusion syndrome by more frequent ultrasound screening may result in improved outcomes for the condition, as fewer pregnancies will be diagnosed in late stages.
Another advantage to earlier detection of twin–twin transfusion syndrome is the ability to effectively counsel patients regarding available management options (expectant management, delivery, termination, selective feticide, septostomy, amnioreduction, laser photocoagulation) and to initiate timely transfer to a treatment center if electing to proceed with intervention. More frequent surveillance of monochorionic pregnancies may allow for earlier detection of other complications of monochorionic gestations such as discordant anomalies, discordant uteroplacental insufficiency, and twin-reversed arterial perfusion sequence.
The objective of any screening strategy is to detect a disease in its early stages, thereby allowing for more timely intervention and, it is hoped, better outcomes. One requirement of a screening strategy is that effective treatment is available for the condition being screened. Given what we currently know about twin–twin transfusion syndrome, this criterion is met in that laser photocoagulation of communicating vessels effectively treats the condition in the majority of patients.11 If detected and managed in its early stages, the diagnosis of twin–twin transfusion syndrome need not be as grim as it has been in the past.
The Maternal-Fetal Medicine division at our institution has adopted a similar strategy to what is recommended in Europe.6,7 Every effort is made to accurately diagnose chorionicity at the first ultrasound examination. We initiate surveillance of known monochorionic gestations no later than 16 weeks and evaluate monochorionic pregnancies every 2 weeks for evidence of twin–twin transfusion syndrome and other complications of monochorionic gestations. Amniotic fluid volume and bladder status are evaluated on each ultrasonogram. Doppler evaluation is preformed only if an abnormality is identified. Growth is measured every 4 weeks. In cases in which findings have raised the suspicion for development of twin–twin transfusion syndrome, patients have increased frequency of ultrasound examinations.
Magee-Womens Hospital is a tertiary care center that serves as an ultrasound referral center for Western Pennsylvania, West Virginia, and Eastern Ohio. Previous ultrasound data were not available on 18 of 42 patients. Judging by the indications for the examinations, these patients were either referred for their anatomic survey or urgently referred because of an abnormal ultrasonogram in the community (suspected twin–twin transfusion syndrome, suspected monoamniotic twins, increased risk for open neural tube defect). When evaluating only the 24 pregnancies with a previous ultrasonogram, our results still indicate that the twin–twin transfusion syndrome is more likely to be detected at a later stage when the ultrasound interval is greater than 14 days (P=.028). This series represents a mix of patients managed by both academic and community ob-gyns. Many referring physicians did not distinguish monochorionic from dichorionic twin gestations, which may explain why half of the cases in our series had “anatomic survey” listed as the indication for the ultrasonogram that diagnosed twin–twin transfusion syndrome, even when monochorionic status was known.
An obvious disadvantage of our study is the retrospective design, which may allow for the introduction of various biases, such as selection and sampling bias. Given the relatively small number of cases, we had to dichotomize variables of interest to short compared with long screening interval and early compared with late stages. The ultrasound screening interval of 14 days or less or greater than 14 days was predetermined based on our current practice. A larger sample size would have allowed for a more thorough analysis of time intervals such as 1 week, 2 weeks, 3 weeks, 4 weeks, and greater than 4 weeks. Nonetheless, only two patients in the 14-days-or-less interval group were diagnosed with late-stage twin–twin transfusion syndrome. A prospective, multicenter cohort of monochorionic-diamniotic gestations would best describe the natural history of these multiple gestations.
In a recent review in the American Journal of Obstetrics and Gynecology, Stamilio and colleagues argue that “… there is normative equipoise to justify the performance of randomized clinical trials to identify the optimal treatment strategy for mild [twin–twin transfusion syndrome].”19 This will require identifying twin–twin transfusion syndrome in the early stages, which occurred in only 57% (24 of 42) of cases in our series.
We encourage obstetric care providers to be more cognizant of the need to treat monochorionic gestations differently than dichorionic gestations, and to recognize that the ultrasound diagnosis of simply “twins” is insufficient. Finally, we hope that our study aids in the development of evidence-based recommendations for sonographic surveillance of monochorionic gestations.
1. Berghella V, Kaufmann M. Natural history of twin-twin transfusion syndrome. J Reprod Med 2001;46:480–4.
2. Sebire NJ, Souka A, Skentou H, Geerts L, Nicolaides KH. Early prediction of severe twin-to-twin transfusion syndrome. Hum Reprod 2000;15:2008–10.
3. Jain V, Fisk NM. The twin-twin transfusion syndrome. Clin Obstet Gynecol 2004;47:181–202.
4. Galea P, Jain V, Fisk NM. Insights into the pathophysiology of twin-twin transfusion syndrome. Prenat Diagn 2005;25:777–85.
5. Quintero RA, Morales WJ, Allen MH, Bornick PW, Johnson PK, Kruger M. Staging of twin-twin transfusion syndrome. J Perinatol 1999;19(8 pt 1):550–5.
6. Management of monochorionic twin pregnancy. Green-top Guideline No. 51. Royal College of Obstetricians and Gynaecologists; 2008:1–13.
7. Twin pregnancies: French 2009 guidelines. National French College in Obstetrics and Gynecology (CNGOF); 2009:1–8.
8. Multiple gestations. American College of Radiology, ACR Appropriateness Criteria. 2008:1–7.
9. AIUM Practice guidelines for the performance of obstetric ultrasound examinations. American Institute of Ultrasound in Medicine. J Ultrasound Med 2010;29:157–66.
10. Multiple gestation: complicated twin, triplet, and high-order multifetal pregnancy. ACOG Practice Bulletin No. 56. American College of Obstetricians and Gynecologists. Obstet Gynecol 2004;104:869–83.
11. Senat MV, Deprest J, Boulvain M, Paupe A, Winer N, Ville Y. Endoscopic laser surgery versus serial amnioreduction for severe twin-to-twin transfusion syndrome. N Engl J Med 2004;352:136–44.
12. Sueters M, Middeldorp JM, Lopriore E, Oepkes D, Kanhai HH, Vandenbussche FP. Timely diagnosis of twin-to-twin transfusion syndrome in monochorionic twin pregnancies by biweekly sonography combined with patient instruction to report onset of symptoms. Ultrasound Obstet Gynecol 2006;28:659–64.
13. Quintero RA, Dickinson JE, Morales WJ, Bornick PW, Bermúdez C, Cincotta R, et al. Stage-based treatment of twin-twin transfusion syndrome. Am J Obstet Gynecol 2003;188:1333–40.
14. Dickinson JE, Evans SF. The progression of disease stage in twin-twin transfusion syndrome. J Matern Fetal Neonatal Med 2004;16:95–101.
15. Duncombe GJ, Dickinson JE, Evans SF. Perinatal characteristics and outcomes of pregnancies complicated by twin-twin transfusion syndrome. Obstet Gynecol 2003;101:1190–6.
16. Crombleholme TM, Shera D, Lee H, Johnson M, D'Alton M, Porter F, et al. A prospective, randomized, multicenter trial of amnioreduction vs selective fetoscopic laser photocoagulation for the treatment of severe twin-twin transfusion syndrome. Am J Obstet Gynecol 2007;396.e1–9.
17. Hack KE, Derks JB, Elias SG, Franx A, Roos EJ, Voerman SK, et al. Increased perinatal mortality and morbidity in monochorionic versus dichorionic twin pregnancies: clinical implications of a large Dutch cohort study. BJOG 2008;115:58–67.
18. Moise KJ Jr, Johnson A. There is NO diagnosis of twins. Am J Obstet Gynecol 2010;203:1–2.
19. Stamilio DM, Fraser WD, Moore TR. Twin-twin transfusion syndrome: an ethics-based and evidence-based argument for clinical research. Am J Obstet Gynecol 2010;203:3–16.