Staging of Twin-Twin Transfusion Syndrome

Original Article n

n

n

n

n

n

n

n

n

n

n

n

n

n

Staging of Twin-Twin Transfusion Syndrome Rubeń A. Quintero, MD Walter J. Morales, MD, PhD Mary H. Allen, RN Patricia W. Bornick, RN, MSN Patricia K. Johnson, RDMS Michael Kruger, MA OBJECTIVE: The purpose of this study was to evaluate the prognostic value of sonographic and clinical parameters to develop a staging classification of twin-twin transfusion syndrome (TTTS). STUDY DESIGN: Severe TTTS was defined as the presence of polyhydramnios (maximum vertical pocket of $8 cm) and oligohydramnios (maximum vertical pocket of #2 cm). Nonvisualization of the bladder in the donor twin (2BDT) and absence of presence of hydrops was also noted. The middle cerebral artery, umbilical artery, ductus venosus, and umbilical vein in both fetuses were assessed with pulsed Doppler. Critically abnormal Doppler studies (CADs) were defined as absent/reverse end-diastolic velocity in the umbilical artery, reverse flow in the ductus venosus, or pulsatile flow in the umbilical vein. TTTS was staged as follows: stage I, BDT still visible; stage II, BDT no longer visible, no CADs; stage III, CADs; stage IV, hydrops; stage V, demise of one or both twins. Laser photocoagulation of communicating vessels (LPCV) or umbilical cord ligation was performed depending on the severity of the condition. The study was approved by the Institutional Review Board of St. Joseph’s Hospital in Tampa and by the Fetal Therapy Board at Hutzel Hospital, Detroit, and all patients gave informed consent. RESULTS: A total of 80 of 108 referred patients met criteria for surgery, but only 65 were treated surgically: 48 with LPCV and 17 with umbilical cord ligation. Complete Doppler data were obtainable in 41 of 48 LPCV patients. Survival rates by stage for one or two fetuses were statistically different (x-squared analysis 5 12.9, df 5 6, p 5 0.044). Neither percent size discordance nor gestational age at diagnosis were predictive of outcome.

Florida Institute for Fetal Diagnosis and Therapy (R. A. Q., W. J. M., M. H. A., P. W. B., P. K. J.), St. Joseph’s Women’s Hospital, Tampa, FL; and Department of Obstetrics and Gynecology (M. K.), Wayne State University, Detroit, MI. Address correspondence and reprint requests to Rubeń A. Quintero, MD, 13601 Bruce B. Downs Boulevard, Suite 160, Tampa, FL 33613.

CONCLUSION: Staging of TTTS using the proposed criteria has prognostic significance. This staging system may allow comparison of outcome data of TTTS with different treatment modalities.

Twin-twin transfusion syndrome (TTTS) occurs in 10% to 17% of monochorionic pregnancies. The syndrome is defined sonographically by the presence of a polyhydramnios in the sac of one twin (recipient twin) and oligohydramnios in the sac of the other twin (donor twin). Usually, marked discordance in the estimated fetal weight is also present, although the former criterion of .20% discordance is no longer used in the diagnosis.1 Other sonographic parameters that aid in the diagnosis include a single placenta, absence of a twin-peak sign, same sex,2,3 hydrops, and abnormal Doppler studies of the arterial or venous territories.4 – 8 Serial ultrasound examinations may show progression of the disease from polyhydramnios/oligohydramnios to frank stuck-twin syndrome, hydrops of either twin, and ultimately intrauterine fetal demise. Despite the known variations in the sonographic presentation of the disease, outcome results have not taken into account such differences. Recently, attempts have been made at identifying specific prognostic factors that could be useful in the counseling of patients. Mari9 and Zikulnig et al.10 have suggested that abnormal Doppler studies may be associated with a poor prognosis. In addition to Doppler data, there is a generalized belief that the earlier the gestational age at presentation and the greater the size discordance, the worse the prognosis. During the last 5 years, we have studied the various sonographic presentations of TTTS. We have noted an apparent time sequence of events as follows: (1) The most benign presentation consists of poly/ oligohydramnios, with the bladder of the donor twin (BDT) still visible. (2) Next, the BDT no longer fills. (3) In more advanced stages, Doppler studies become severely abnormal; typically the Donor twin will show arterial abnormalities, with absent end-diastolic velocity (AEDV) in the umbilical artery (UA), or the recipient twin will show venous abnormalities such as reverse flow in the ductus venosus (RFDV) or pulsatile umbilical venous flow (PUVF), usually associated with marked tricuspid regurgitation; (4) Eventually, either twin becomes hydropic. (5) Finally, fetal demise of either or both twins occurs. The purpose of this article was to assess the possible prognostic significance of a staging system that would include sonographic and pulsed Doppler data. STUDY DESIGN Patients referred with the diagnosis of TTTS were included in the study. Patients referred for possible in utero surgery underwent a Journal of Perinatology (1999) 19(8) Part 1, 550 –555 © 1999 Stockton Press. All rights reserved. 0743– 8346/99 $12

550

http://www.stockton-press.co.uk

Staging of TTTS

Quintero et al.

Table 2 Nonsurgical Patients With Confirmed TTTS (N 5 15)

Table 1 Staging of TTTS Based on Sonographic and Doppler Findings Stage

I II III IV V

Poly/ oligohydramnios*

Absent bladder in donor

CADs†

1 1 1 1 1

2 1 1 1 1

2 2 1 1 1

Hydrops

2 2 2 1 1

2 2 2 2 1

*Polyhydramnios, MVP of .8 cm; oligohydramnios, MVP of ,2 cm. †CADs, defined as the presence of at least one of the following: (1) UA AEDV/REDV, (2) RFDV, or (3) pulsatile umbilical venous flow (PUVF).

comprehensive ultrasound examination that included fetal anatomy, biometry, maximum vertical pocket (MVP) of amniotic fluid in each sac, and presence or absence of hydrops. The ultrasound assessment lasted ;60 minutes. Doppler waveform analysis included the middle cerebral artery, umbilical artery, ductus venosus, and umbilical vein. The presence or absence of tricuspid regurgitation was also noted. TTTS was defined as polyhydramnios (MVP of .8 cm) in the recipient twin and oligohydramnios (MVP of ,2 cm) in the Donor twin. In addition, the presence of a single placenta, absent twin peak sign, and same sex were present in all cases. Patients with confirmed TTTS were staged as follows (Table 1): stage I, polyhydramnios in recipient twin, donor twin with severe oligohydramnios but with visible BDT; stage II, polyhydramnios in recipient twin, donor twin stuck, BDT not visible at any point during the preoperative evaluation, including during the ultrasound performed by the referring physician, diastolic flow present in the umbilical artery and forward flow in the ductus venosus; stage III, poly/oligohydramnios, BDT not visible, and critically abnormal Doppler studies (CADs) defined by at least one of the following: (1) AEDV or reverse end-diastolic velocity in the umbilical artery (REDV), (2) RFDV, or (3) PUVF. Patients with CADs but with the BDT still visible were also classified as stage III. Stage IV, presence of ascites or frank hydrops (fluid collection in two or more cavities) in either fetus; stage V, demise of either fetus. In general, patients in stage I were not offered surgical treatment and were managed expectantly or with serial amniocentesis if symptoms of preterm labor or severe abdominal distension were present. However, some patients with stage I disease were treated with laser early in our series. Patients with stage II disease were offered laser photocoagulation of communicating vessels (LPCV). From August of 1993 to June of 1997, we used the technique described by Ville et al11 of lasering all vessels crossing the dividing membrane. Since July of 1997, we have used a selective technique developed by us (selective LPCV) that interrupts only the communicating vessels (deep or superficial communications), while sparing those vessels that do not participate in blood exchange between the fetuses, regardless of the location of the vessels relative to the dividing membrane.12 Briefly, in this Journal of Perinatology (1999) 19(8) Part 1, 550 –555

Status

N

Lost to follow-up Termination of pregnancy Serial amniocentesis Miscarried, intrauterine fetal demise At least one survivor

4 1 10 4 6

Demise

technique all vessels are followed to their terminal end in the placenta; arteries with a returning vein to the same fetus represent normally perfused placental areas, regardless of whether the vessels cross the membrane. On the contrary, arteries without a returning vein to the same fetus, but with a vein going to the other twin, represent deep arterio-venous anastomoses. Superficial arterio-arterial or venovenous anastomoses are easily identified, as neither of these vessels have a terminal end in the placenta, but rather continue uninterrupted from one cord to the other.12 Most patients with stage III disease were offered LPCV, although some underwent umbilical cord ligation (UCL) for severe disease at early presentation.13 Patients with stage IV disease were primarily offered UCL. Patients in stage V were offered no treatment, as any adverse consequences of spontaneous fetal demise on the surviving twin could no longer be prevented. Gestational age at presentation and size discordance were also analyzed as possible prognostic factors. Patients undergoing surgery or amniocentesis at our center were asked to give informed consent. The surgical protocols were approved by the Fetal Therapy Board of Hutzel Hospital (Wayne State University, Detroit, MI) and the Institutional Review Board of St. Joseph’s Hospital (Tampa, FL). Statistical Analysis Statistical analysis was conducted using SPSS version 8.0 for Windows 95. Categorical variables such as stage and survival were analyzed using the x-squared analysis and Fisher’s exact test. Interval variables such as gestational age and estimated fetal weight discordance were analyzed with Student’s t-test and one-way analysis of variance with Student-Newman-Keuls post-hoc tests. Spearman’s correlation coefficient was used to analyze linear by linear associations where needed. RESULTS A total of 108 consultations of patients with the referring diagnosis of TTTS were received between August of 1993 and January of 1999. Of these, 80 patients met the criteria for TTTS and 65 patients were treated surgically. Of the 15 patients that were not treated surgically, 10 underwent serial amniocentesis by their referring physicians or by us, 4 were lost to follow-up, and 1 terminated the pregnancy (Table 2). Of the 10 patients treated with amniocentesis, 5 had dual survivors and 1 had a single survivor, 3 miscarried, and 1 had an intrauterine demise. Nine of the 10 patients treated with amniocentesis could be 551

Quintero et al.

Staging of TTTS

Table 3 Survival Rates by Stage in Patients Treated With Amniocentesis: Zero, One, or Two Survivors (N 5 9/10 with complete data) Stage

Zero survivors N (%)

One survivor N (%)

Two survivors N (%)

Total

I II III Total

0 (0%) 1 (33%) 2 (66%) 3 (33%)

1 (100%) 0 (0%) 0 (0%) 1 (11%)

5 (100%) 0 (0%) 0 (0%) 5 (56%)

6 1 2 9

staged retrospectively. All patients with a successful outcome were in stage I (Table 3). A total of 65 patients were treated surgically: 48 underwent LPCV and 17 underwent UCL. Of the 48 patients treated with LPCV, at least one fetus survived in 65% of cases (31 of 48), but complete Doppler data were available in both twins only in 41 cases. Tables 4 and 5 show the outcome of patients treated with LPCV by stage. At least one fetus survived in 3 of 3 (100%) patients with stage I disease, in 10 of 13 (77%) patients with stage II disease, and in 12 of 18 (67%) patients with stage III disease. In stage IV, at least one fetus survived in 3 of 7 (43%) pregnancies. A statistically significant difference in the survival of zero, one, or two fetuses was found between stages (x-squared analysis 5 12.9, df 5 6, p 5 0.044) (Table 4). There were more single survivors in stage II (9 of 13, 69%) than in stage III (4 of 18, 22%), accounting for the difference between these two stages (xsquared analysis 5 7.8, df 5 2, p 5 0.021). A negative correlation was found between survival of at least one fetus and stage (Table 5) (Spearman’s correlation coefficient 5 20.29, p 5 0.03, one-tailed analysis). Of the 12 stage III patients with a successful outcome, 4 had a single fetal survivor; in two of the four, the surviving twin was the fetus with critically abnormal Dopplers. CADs were not predictive of survival in the donor twin ( p 5 0.34, Fisher’s exact test, one-tailed analysis) or in the recipient twin ( p 5 0.10, Fisher’s exact test, onetailed analysis). A total of 17 patients were treated with UCL (Table 6). Complete Doppler data were available for both twins in 15 cases. Two of 2 patients in stage III (100%) and 10 of 13 patients in stage IV (77%) had a successful outcome. There were no differences in outcome between these two stages in patients treated with UCL ( p 5 0.63, one-tailed analysis, Fisher’s exact test). The prognostic value of gestational age at the time of presentation and the degree of size discordance were also assessed. Patients were grouped as presenting before or after 20 weeks’ gestation (Tables 7 and 8). Outcomes were not statistically different between the two groups when analyzed either by zero, one, or two survivors or by at least one survivor. In fact, patients treated at ,20 weeks appeared to have a higher success rate than those treated between 20 and 26 weeks (11 of 14 (79%) vs. 20 of 34 (59%)), but this difference was not statistically significant ( p 5 0.17, Fisher’s exact test, one-tailed analysis). Estimated fetal weight percent discordance was classified into three 552

Table 3B Survival Rates by Stage in Patients Treated With Amniocentesis: At Least One Survivor (N 5 9/10 with complete data) Stage


At least one survivor N (%)

Total

I II III Total

0 (0%) 1 (33%) 2 (66%) 3 (33%)

6 (66%) 0 (0%) 0 (0%) 6 (66%)

6 1 2 9

Table 4 Survival Rates in Patients Treated With LPCV: Zero, One, or Two Survivors (N 5 41/48 patients with complete Doppler data)* Stage


One survivor N (%)

Two survivors N (%)

Total

I II III IV Total

0 (0%) 3 (23%) 6 (28%) 4 (57%) 13 (32%)

1 (33%) 9 (69%) 4 (22%) 1 (14%) 15 (36%)

2 (67%) 1 (8%) 8 (45%) 2 (29%) 13 (32%)

3 13 18 7 41

*x-squared analysis 5 12.9, df 5 6, p 5 0.044.

groups: ,20%, between 20% and 40%, and .40%. Theoretically, the greater the size discordance, the more likely the donor twin would not survive. Table 9 shows that size discordance was not predictive of survival of the donor twin, with survival rates of 50% of the donor twin even in patients with .40% discordance (x-squared analysis 5 3.4, df 5 2, p 5 0.09, one-tailed analysis). Overall, 25 of 48 (52%) recipient twins and 20 of 48 (41%) donor twins treated with LPCV survived. In patients with a single survivor (N 5 17), there were almost twice as many recipient fetuses than donors (11 vs. 6), but this difference was not statistically significant (binomial test, p 5 0.33). DISCUSSION The natural history of TTTS is variable. Although not all cases of TTTS may show progressive deterioration, time-related changes may occur if the disease is managed expectantly or symptomatically. In cases with progressive disease, we have noticed an apparent sequence of events, starting from polyhydramnios/oligohydramnios, followed by nonvisualization of the BDT, development of CADs, ascites, or frank hydrops, and demise (Figure 1). We call this “the spectrum of TTTS.” Whether a particular patient will show stable signs of TTTS or rapidly progressive disease is currently unpredictable. Because the disease can present at different stages, analysis of treatment outcomes must take into account not only the type of treatment offered but also the extent of the disease at the time of treatment. Our data show that assessing the degree of severity of the disease with the current staging system may be of prognostic value. Journal of Perinatology (1999) 19(8) Part 1, 550 –555

Staging of TTTS

Quintero et al.

Table 5 Survival Rates in Patients Treated With LPCV: At Least One Survivor (N 5 41/48 patients with complete Doppler data)*

Table 6 Survival Rates for TTTS Stages III–IV Treated With UCL (N 5 15/17 with complete Doppler data)*

Stage



Total

Stage


One survivor N (%)

Total

I II III IV Total

0 (0%) 3 (23%) 6 (33%) 4 (57%) 13 (32%)

3 (100%) 10 (77%) 12 (67%) 3 (43%) 28 (68%)

3 13 18 7 41

III IV Total

0 (0%) 3 (23%) 3 (20%)

2 (100%) 10 (77%) 12 (80%)

2 13 15

*p 5 0.63, Fisher’s exact test, one-tailed analysis.

*Spearman’s correlation coefficient: 20.29, p 5 0.03 one-tailed analysis.

Table 7 Survival Rates by Gestational Age in Patients Treated With LPCV: Zero, One, or Two Survivors (N 5 48)*

Non-stuck TTTS, or stage I disease, is characterized by polyhydramnios/oligohydramnios, but the BDT is still visible and the Doppler studies are not critically abnormal. It is important to differentiate stage I disease from simple amniotic fluid volume discrepancy. In the latter, although the fluid volumes may be different, there is no polyhydramnios. Non-TTTS amniotic fluid discrepancies are known to occur in ;26% of monochorionic twins (Nicolaides, personal communication). Our reason for offering mostly amniocentesis and not laser therapy in stage I disease is based on the assumption that the blood volume imbalance between the fetuses is probably not severe enough to warrant a more definitive therapy, and that in fact, many of these patients may not show progressive disease. The presence of a visible BDT suggests that renal perfusion is preserved in this fetus, and that the blood volume lost to the recipient twin is probably not sufficiently large to result in demise of the donor twin at that stage. Although the number of patients treated with amniocentesis was small, the 66% (6 of 9) survival rate observed in these patients suggests that, indeed, patients with less severe disease may be managed with amniocentesis provided that the condition does not progress to more advanced stages. Indeed, all of the survivors in the amniocentesis group corresponded to stage I. Comparison with 100% (3 of 3) survival rate in stage I patients treated with LPCV is not possible due to the small number of patients. Conversely, patients presenting with more advanced disease may require more definitive therapy (LPCV or UCL). Comparison of amniocentesis versus LPCV outcomes, excluding stage I patients, showed a trend toward improved outcomes with LPCV in more severe states (0 of 3 survived with amniocentesis versus 25 of 38, p 5 0.053, Fisher’s exact test). In our practice, patients with stage I disease are seen weekly for possible progression of the disease. If the disease progresses to stage II to IV, patients are then offered surgery. Stage II and stage III disease correspond to the classic “stucktwin syndrome,” with characteristic lack of visualization of the BDT and polyhydramnios in the recipient twin. The sonographic image of stuck twin resulting from TTTS should be differentiated from similar images as in bilateral renal agenesis, undiagnosed premature rupture of membranes, or severe intrauterine growth retardation. In these other disorders, polyhydramnios of the other twin is not present. Absence of bladder filling is an important diagnostic criterion in stage II and III because, although a small amount of fluid may still be Journal of Perinatology (1999) 19(8) Part 1, 550 –555

Gestational age


One survivor N (%)

Two survivors N (%)

Total

,20 weeks $20 weeks

3 (21%) 14 (38%)

7 (50%) 10 (29%)

4 (28%) 10 (32%)

14 34

*x-squared analysis 5 2.28, df 5 2, p 5 0.16, one-tailed analysis.

Table 8 Survival Rates by Gestational Age in Patients Treated With LPCV: At Least One Survival* Gestational age



Total

,20 weeks $20 weeks Total

3 (21%) 14 (38%) 17 (34%)

11 (79%) 20 (59%) 31 (65%)

14 34 48

*p 5 0.17, Fisher’s exact test, one-tailed analysis.

present in the sac of the donor, this fluid is unlikely to result from fetal urination but rather from placental, umbilical cord, or fetal skin sources. Patients in whom the amniotic fluid volume of the donor twin normalizes after amniocentesis but with 2BDT should also be classified as “stuck.” In these patients, unintentional piercing of the dividing membrane has probably occurred, allowing fluid from the recipient twin to enter the sac of the donor twin. Obviously, purposeful disruption of the dividing membrane may have similar results. Filling of the BDT is also an important sonographic parameter to monitor the effect of therapy. In our experience, bladder filling may return from 5 hours to 5 days after surgery, reflecting improved renal perfusion of the donor twin. The criterion to differentiate stages II and III is the absence or presence of CADs. Typically, stage III patients will also have a 2BDT, but rarely the donor twin may still have a visible bladder; in these cases, patients are still classified as stage III (atypical stage III). Zikulnig et al.10 have shown that RFDV is associated with an increased likelihood of demise of the recipient twin, but could not show a similar association with UA AEDV/REDV. Typically, the recipient twin is 553

Quintero et al.

Staging of TTTS

Table 9 Survival Rate of the Donor Twin by Estimated Fetal Weight Discordance in Patients Treated With LPCV (N 5 45/48 with complete estimated fetal weight data)* Estimated fetal weight % discordance

Donor survivor

Donor nonsurvivor

Total

0–20% 21–40% 41–60% Total

7 (47%) 15 (75%) 5 (50%) 27 (60%)

8 (53%) 5 (25%) 5 (50%) 18 (40%)

15 (33%) 20 (44%) 10 (22%) 45 (100%)

*x-squared analysis 5 3.4, df 5 2, p 5 0.09, one-tailed analysis.

more likely to show abnormal venous flow studies, whereas the donor twin may show deranged umbilical artery studies;7 however, either fetus may show alterations in one or both systems. Our data did not show UA AEDV/REDV or RFDV/PUVF alone to be predictive of the outcome of the fetus with CADs, but the number of single survivors in this subgroup was too small (four fetuses). However, when analyzed by stage, stage III patients had a statistically different outcome than stage II patients (x-squared analysis 5 7.8, df 5 2, p 5 0.02). Because our staging system combines arterial and venous abnormal blood flow information for staging, grouping patients with CAD into a single stage may better reflect a similar degree of disease in these patients, regardless of whether the CAD is present in the donor or the recipient twin. Importantly, because CADs were not predictive of the survival of the non-CAD fetus, identification of the CAD fetus may not be useful in determining which fetus is more likely to survive. There were more dual survivors in stage III than in stage II. Although this may seem paradoxical, we have learned that the survival of one versus two fetuses may not depend on the operator (except when using nonselective LPCV). Instead, survival of one or two fetuses may reflect differences in placentation, which unfortunately cannot be diagnosed endoscopically. The presence of hydrops (stage IV) is unquestionably a poor prognostic sign. However, we could not find a statistically significant difference between stages III and IV. Perhaps this could be due to the small number of patients in this category (a total of 42 cases would have been necessary to show a difference). Five of 7 patients were treated with non-selective LPCV (before 1997), with 2 of 5 (40%) having at least one surviving fetus. Due to these unfavorable results, and before developing selective LPCV, we offered UCL as a first line of treatment to patients with stage IV presenting before 24 weeks’ gestation. In these patients, the survival rate with UCL was 77% (10 of 13) (Table 6). Recently, we performed selective LPCV in a patient referred at 25 weeks’ gestation after three previous amniocenteses. The recipient twin had ascites, bilateral pleural effusion, and scalp and skin edema. UCL was not offered due to the advanced gestational age. Selective LPCV resulted in disappearance of the scalp edema and the effusions, and only minimal ascites remained. The patient was delivered at 31 weeks’ gestation, and both babies survived. Although this case illustrates the potential for selective LPCV to revert hydrops, this 554

Figure 1 Proposed sequence of events in progressive TTTS: Polyhydramnios/oligohydramnios is followed by nonvisualization of the BDT, development of CADs, ascites, or frank hydrops, and demise.

should be viewed with caution, as the prognosis for these fetuses in terms of cardiac or neurologic function is unknown. Therefore, in previable stage IV disease, UCL of the hydropic fetus should still be viewed as the preferred treatment option. Analysis of our data did not show gestational age alone to be predictive of outcome. This is in contrast to the general belief that the earlier the gestational age the poorer the prognosis. In fact, the survival rate was apparently lower in patients diagnosed after 20 weeks (79% vs. 59%, p 5 0.17, Fisher’s exact test). This finding may reflect the attitude of physicians to refer patients promptly for laser therapy when TTTS is diagnosed early in gestation, anticipating having to perform a significant number of amniocenteses. In turn, a lower survival rate in patients treated after 20 weeks may reflect more advanced disease and possibly referral delay. Indeed, 76% of patients (19 of 25) with stage III or IV disease were treated after 20 weeks, and all stage IV patients were .20 weeks at the time of treatment. Although size discordance should not be used as a criterion for the diagnosis of TTTS, it may have pathophysiological importance. Significant size discordance may suggest the possibility that two separate pathologic mechanisms, selective intrauterine growth retardation from disparate placental distribution or TTTS, may be acting concomitantly. Because the percentage of size discordance was not predictive of survival for both twins or for the donor twin alone, important size discrepancies should not necessarily be viewed as a poor prognostic sign. In general, however, the ratio of recipient/donor survival was ;2:1 (11 recipients versus 6 donors) (patients treated with LPCV with a single survivor), suggesting an overall general relative disadvantage of the donor twin. There were several limitations in our study. First, staging was assigned retrospectively, resulting in incomplete data in several patients. Second, in addition to UCL and amniocentesis, patients were treated with two different laser surgical techniques, namely nonselective LPCV (before July 1997) or selective LPCV. In this regard, the survival analysis of patients treated with LPCV (Tables 4 and 5) should be viewed with caution, as we expect that survival statistics will Journal of Perinatology (1999) 19(8) Part 1, 550 –555

Staging of TTTS

Quintero et al.

improve with the new selective technique. Finally, the number of patients is relatively small when broken down into four or more different categories, making statistical analysis difficult. Nonetheless, and despite these limitations, it is our impression that the differences in the severity of the disease as gauged by our proposed staging system seem to have prognostic value. We hope that the current staging system can be useful in the counseling of patients, in the individualization of therapeutic options, and in the comparison of outcome results between centers. In addition, we hope that anatomical, biometrical, and Doppler data, as suggested by us, may be incorporated into the standard sonographic evaluation of patients with TTTS. Further experience will determine the advantages and limitations of the proposed staging system for TTTS.

3. Ville Y. Monochorionic twin pregnancies: “les liaisons dangereuses.” Ultrasound Obstet Gynecol 1997;10:82–5.

Acknowledgments

9. Mari G. International TTTS registry group. Am J Obstet Gynecol 1998:s28.

We thank all referring physicians; our colleagues Dr. Jeffrey L. Angel, Dr. Craig S. Kalter, and Dr. Gregg Giannina; our sonographers Connie Rasmussen, Karen Pomeroy, and Kim Valdez; our nurses Lisa Farrell, Debbie Neal, Cheri Clause, and Katy Papageorge; our nurse anesthetist Ruth Neumann and Universal Anesthesia Care, Inc.; our research assistant Irene S. Quintero; and the staff at Florida Perinatal Associates for their collaboration. We also extend our gratitude to Richard Wolf Inc., Surgical Laser Technologies, and Cook Ob/Gyn for their technical support.

References 1. Danskin FH, Neilson JP. Twin-to-twin transfusion syndrome: what are appropriate diagnostic criteria? Am J Obstet Gynecol 1989;161:365–9. 2. Wittman B, Baldwin V, Nichol B. Antenatal diagnosis of twin transfusion syndrome by ultrasound. Obstet Gynecol 1981;58:123–7.

Journal of Perinatology (1999) 19(8) Part 1, 550 –555

4. Rizzo G, Arduini D, Romanini C. Cardiac and extracardiac flows in discordant twins. Am J Obstet Gynecol 1994;170:1321–7. 5. Ishimatsu J, Yoshimura O, Manabe A, Matsuzaki T, Tanabe R, Hamada T. Ultrasonography and Doppler studies in twin-to-twin transfusion syndrome. Asia Oceania J Obstet Gynaecol 1992;18:325–31. 6. Ohno Y, Ando H, Tanamura A, Kurauchi O, Mizutani S, Tomoda Y. The value of Doppler ultrasound in the diagnosis and management of twin-to-twin transfusion syndrome. Arch Gynecol Obstet 1994;255:37– 42. 7. Hecher K, Ville Y, Snijders R, Nicolaides K. Doppler studies of the fetal circulation in twin-twin transfusion syndrome. Ultrasound Obstet Gynecol 1995;5:318 –24. 8. Hecher K, Ville Y, Nicolaides KH. Fetal arterial Doppler studies in twin-twin transfusion syndrome. J Ultrasound Med 1995;14:101– 8. 10. Zikulnig L, Hecher K, Bregenzer T, Ba¨z E, Hackeloër B. Doppler sonography and prognostic factors in severe twin-twin transfusion syndrome with endoscopic laser therapy. Ultrasound Obstet Gynecol 1998;12:9 (Abstract). 11. Ville Y, Hyett J, Hecher K, Nicolaides K. Preliminary experience with endoscopic laser surgery for severe twin-twin transfusion syndrome. N Engl J Med 1995;332: 224 –7. 12. Quintero R, Morales W, Mendoza G, et al. Selective photocoagulation of placental vessels in twin-twin transfusion syndrome: evolution of a surgical technique. Obstet Gynecol Surv 1998;53:s97–s103. 13. Quintero RA, Romero R, Reich H, et al. In utero percutaneous umbilical cord ligation in the management of complicated monochorionic multiple gestations. Ultrasound Obstet Gynecol 1996;8:16 –22.

555