Percent absent end-diastolic velocity in the ... - Wiley Online Library

23 downloads 6538 Views 264KB Size Report
Ultrasound Obstet Gynecol 2007; 30: 35–39 ... Hospital, Tampa, Florida and †Obstetrics and Gynecology, Keck School of Medicine, University Southern ...
Ultrasound Obstet Gynecol 2007; 30: 35–39 Published online in Wiley InterScience (www.interscience.wiley.com). DOI: 10.1002/uog.4055

Percent absent end-diastolic velocity in the umbilical artery waveform as a predictor of intrauterine fetal demise of the donor twin after selective laser photocoagulation of communicating vessels in twin–twin transfusion syndrome E. V. KONTOPOULOS*, R. A. QUINTERO*, R. H. CHMAIT†, P. W. BORNICK‡, Z. RUSSELL* and M. H. ALLEN‡ *Department of Obstetrics and Gynecology, Division of Maternal-Fetal Medicine, University of South Florida and ‡Tampa General Hospital, Tampa, Florida and †Obstetrics and Gynecology, Keck School of Medicine, University Southern California, Los Angeles, California, USA

K E Y W O R D S: absent end-diastolic flow; donor twin; Doppler; intrauterine fetal demise; twin–twin transfusion syndrome

ABSTRACT Objective Absent end-diastolic velocity (AEDV) in the umbilical artery of the donor twin is a known risk factor for intrauterine fetal demise (IUFD) of this fetus after selective laser photocoagulation of communicating vessels (SLPCV) for twin–twin transfusion syndrome (TTTS). The aim of this study was to assess the proportion of time, expressed as a percentage, of the cardiac cycle spent in AEDV (%AEDV) as a predictor of IUFD of the donor. Methods All patients referred for possible SLPCV underwent complete preoperative staging evaluation including Doppler assessment of the umbilical artery. %AEDV was calculated retrospectively as 100 × (time of the cycle spent in AEDV divided by duration of total cardiac cycle). Patients without AEDV were considered to have a %AEDV of 0. Follow-up Doppler studies were performed 16–24 h after SLPCV. IUFD of the donor was recorded if the donor twin died any time prior to delivery. Results Of 401 patients undergoing SLPCV, 127 had AEDV. Preoperative AEDV of the donor twin was associated with an increased risk of IUFD of the donor (40.9% vs. 14.2%, P < 0.0001). %AEDV was measured in 72/127 (56.7%) donors with AEDV for whom digital images were available. Within these 72 patients, the mean %AEDV was significantly higher in patients with IUFD of the donor (36.5% vs. 29.6%, P = 0.01). IUFD of the donor was similar in patients with AEDV, regardless of whether %AEDV was measured (36% vs. 47%, P = 0.2). A %AEDV > 30 was associated with a 4.3-fold increase

in the risk of IUFD of the donor (95% CI, 1.4–12.7), a sensitivity of 77% and a negative predictive value of 81.3%. Logistic regression showed that %AEDV, but not number of anastomoses, placental location, presence of artery-to-artery anastomoses or the presence or absence of EDV was associated significantly with IUFD of the donor. Conclusion %AEDV is a novel Doppler parameter in the assessment of patients with TTTS. %AEDV, rather than AEDV alone, is a significant risk factor for IUFD of the donor twin and %AEDV > 30 is associated with an increased risk of IUFD of the donor in TTTS patients treated with SLPCV. Assessment of %AEDV should be considered part of the preoperative evaluation of TTTS patients. Copyright  2007 ISUOG. Published by John Wiley & Sons, Ltd.

INTRODUCTION Twin–twin transfusion syndrome (TTTS) is thought to result from a net transfer of blood from the donor twin to the recipient twin through placental vascular anastomoses. In singleton gestations, absent end-diastolic velocity (AEDV) in the Doppler waveform of the umbilical artery has been associated with poor perinatal outcome1 – 3 , and in twins it is used in the sonographic staging of TTTS; it is typically seen in the donor twin in Stage-III TTTS (Stage-III donor)4 . Preoperative AEDV is a known poor prognostic factor in TTTS5,6 and it is a risk

Correspondence to: Dr E. V. Kontopoulos, Maternal Fetal Medicine, University of South Florida, 4 Columbia Dr., HMT #510, Tampa, FL 33606, USA (e-mail: [email protected]) Accepted: 16 April 2007

Copyright  2007 ISUOG. Published by John Wiley & Sons, Ltd.

ORIGINAL PAPER

Kontopoulos et al.

36

factor for in-utero fetal demise (IUFD) of the donor twin after selective laser photocoagulation of communicating vessels (SLPCV)7 – 9 . By definition, AEDV is a categorical variable. However, we noted that the gestalt appearance of AEDV varied between patients undergoing SLPCV (Figure 1), and that this could be described by calculating the proportion of time, expressed as a percentage, of the cardiac cycle spent in AEDV (%AEDV) (Figure 2). The purpose of this study was to assess the potential association between %AEDV and IUFD of the donor twin in patients with TTTS treated with SLPCV.

METHODS From August 2001 to August 2005, a comprehensive preoperative ultrasound examination was performed on all patients who were referred to our institution with a

Figure 1 Doppler waveform analysis of the umbilical artery in two different patients. The overall appearance of the diastolic time period differs between them, with more absent diastolic flow in (b).

Copyright  2007 ISUOG. Published by John Wiley & Sons, Ltd.

%AEDV = 100 × AEDV / cycle

AEDV

Cardiac cycle

Figure 2 Percent absent end-diastolic velocity (%AEDV) is calculated as: 100 × (length of time during the cardiac cycle spent in (AEDV + REDV)/duration of the total cardiac cycle).

diagnosis of TTTS between 16 and 26 weeks’ gestation. Triplet and higher-order pregnancies and monoamniotic twin pregnancies were excluded from this analysis. Fetal anatomy and biometry, the maximum vertical pocket (MVP) in each sac, cervical length and Doppler studies of the umbilical artery, umbilical vein, ductus venosus, middle cerebral artery and tricuspid valve were obtained. TTTS was diagnosed if the MVP was ≤ 2 cm in the donor’s gestational sac and ≥ 8 cm in the recipient’s sac. Severity of the disease was then classified according to the Quintero staging system4 (Stage I: bladder of the donor twin visible, normal Dopplers; Stage II: bladder of the donor twin not visible, normal Dopplers; Stage III: critically abnormal arterial or venous Doppler in either twin; Stage IV: hydrops fetalis; Stage V: demise of one or both twins). For Doppler examination, the umbilical artery was sampled in a free loop of the umbilical cord. The angle of insonation was kept at < 30◦ . The wall motion filter was kept at < 100 MHz and Doppler energy output levels were < 50 mW/cm2 . Doppler velocimetry examinations were performed in the absence of fetal breathing. AEDV was defined as absent diastolic flow during a typical 6-s window of observation. %AEDV was calculated as 100 × length of time during the cardiac cycle spent in AEDV/duration of the total cardiac cycle. Measurements were obtained retrospectively by digital analysis of available images. At least three waveforms were assessed and the smallest value was used for analysis. Patients with intermittent AEDV were considered to have positive diastolic velocity. %AEDV in patients with reversed end-diastolic velocity (REDV) was calculated as 100 × (length of time during the cardiac cycle spent in (AEDV + REDV)/duration of the total cardiac cycle). Patients without AEDV were considered to have %AEDV = 0. Intra- and interobserver variability was assessed by having two different observers calculate %AEDV in 10 arbitrary samples. SLPCV surgery was performed percutaneously as described previously10 . Prior to 2002 it was performed with the patient under general anesthetic, and from 2002 onwards local anesthetic with intravenous conscious sedation was used. Patients underwent a follow-up ultrasound examination 16–24 h after surgery. Postoperative ultrasound

Ultrasound Obstet Gynecol 2007; 30: 35–39.

Percent absent end-diastolic velocity in TTTS assessment included documentation of fetal heart rate, assessment of MVPs, repeat Doppler studies of the middle cerebral artery, umbilical vessels and ductus venosus, cervical length measurement and we noted if there was evidence of membrane detachment. Patients were typically discharged on postoperative day 1. Weekly follow-up ultrasound examinations were arranged through the referring physician for 1 month, with monthly ultrasound examinations thereafter. Perinatal outcome was obtained through preprinted outcome sheets sent to the referring physicians. Data collected included indication for delivery, mode of delivery, birth weight, Apgar scores and details of any neonatal complications. If both fetuses were liveborn, arrangements were made to collect the fresh placentae in triple-sealed containers. Placentae were assessed at our hospital pathology department as described previously11 . Patency of vascular anastomoses was assessed by air injection8 . Analyses were conducted using SPSS version 13.0 (SPSS Inc, Chicago, IL, USA). All continuous variables were assessed for normality and compared using Student’s ttest or the Mann–Whitney U-test and categorical data were analyzed by chi-square test or Fisher’s exact test, as appropriate. Stepwise logistic regression was performed to assess the contribution of %AEDV to IUFD of the donor. A P-value of < 0.05 was considered statistically significant.

RESULTS Four-hundred and one patients who met the study criteria underwent SLPCV during the study period. Of these, 127 (31.7%) were found to have preoperative AEDV of the donor twin on Doppler evaluation of the umbilical artery. There were no patients with REDV. Overall, 91/401 (22.7%) patients experienced IUFD of the donor twin. The incidence of IUFD of the donor twin was significantly higher in patients with AEDV compared with those without AEDV (52/127, 40.9% vs. 39/274, 14.2%, P < 0.0001). There were nine patients with intermittent AEDV. None of these patients experienced IUFD of the donor twin. Table 1 shows the demographic characteristics of the population relative to the outcome of IUFD of the donor twin. Although gestational age at the time of surgery and operating time were statistically different between the two groups, these differences are unlikely to be of clinical significance. The groups were not different in terms of gestational age at delivery, preoperative cervical length, number of anastomoses or incidence of anterior placenta. Patients with IUFD of the donor twin were more likely to be Stage-III and less likely to be Stage-II or Stage-IV TTTS. This is in agreement with the finding that AEDV of the umbilical artery of the donor is associated with a higher likelihood of IUFD of the donor. Images were available for review and analysis in 72/127 (56.7%) patients with preoperative AEDV of the donor twin’s umbilical artery. The intraclass correlation coefficient for intra- and interobserver assessments of %AEDV

Copyright  2007 ISUOG. Published by John Wiley & Sons, Ltd.

37 Table 1 Clinical characteristics of the patient population

Characteristic

Donor survived Donor died (n = 310; 77.3%) (n = 91; 22.7%)

GA at surgery 20.3 (16–26) (weeks) GA at delivery 33.1 (18–40) (weeks) Operating time 40 (17–188) (min) Anterior placenta 144 (46.5) Cervical length (cm) 4 (0.8–6.2) Number of 6 (1–15) anastomoses per patient 52 (16.8) Arterioarterial anastomoses* TTTS Stage I 48 (15.5) TTTS Stage II 93 (30) TTTS Stage III 130 (41.9) TTTS Stage IV 39 (12.6)

P

19.7 (16–26)

0.003

33.6 (18–40)

0.7

48 (13–144)

0.003

39 (42.9) 4 (1.4–5.8) 5.5 (1–18)

0.5 0.09 0.8

28 (30.8)

0.003

8 (8.8) 13 (14.3) 66 (72.5) 4 (4.4)

0.1 0.002 < 0.001 0.02

Values are given as n (%) or median (range). *Number of patients in group with at least one arterioarterial anastomosis. GA, gestational age.

were 98% (95% CI, 92.3–99.5) and 99% (95% CI, 98–99), respectively. Twenty-six of these 72 (36.1%) patients experienced IUFD of the donor twin. The mean %AEDV was higher in patients with compared with those without IUFD of the donor twin (36.5 ± 10.9% vs. 29.6 ± 10.3%, P = 0.01). The incidence of IUFD of the donor twin was similar in patients with AEDV, regardless of whether %AEDV was measured or not (36% vs. 47%, P = 0.2). A receiver–operating characteristics curve was constructed to determine the optimal cut-off point of %AEDV for the prediction of IUFD of the donor (Figure 3) (area under the curve: 0.68; 95% CI, 0.55–0.81; P = 0.01). A %AEDV > 30 was associated with 4.3-fold (95% CI, 1.4–12.7) increase in the likelihood of IUFD of the donor, with a sensitivity of 76.9% and a negative predictive value of 81.3%. Of the 72 patients with a %AEDV value, 40 (55.6%) had %AEDV > 30. Stepwise logistic regression was performed to discriminate between the contribution of AEDV vs. that of %AEDV and other factors associated with IUFD of the donor. Logistic regression showed IUFD of the donor twin to be associated positively with %AEDV (P < 0.001; odds ratio (OR), 1.04; 95% CI, 1.02–1.06) and inversely with gestational age at the time of surgery (P = 0.003; OR, 0.79; 95% CI, 0.67–0.92) as well as with operating time (P = 0.03; OR, 1.01; 95% CI, 1.0–1.02). The mere presence or absence of AEDV was not associated significantly with IUFD of the donor twin (P = 0.5). Logistic regression including %AEDV > 30 showed a strong association with IUFD of the donor twin (P < 0.001; OR, 6.8; 95% CI, 3.3–13.8), with gestational age at the time of the procedure and operating time remaining in the equation. Number of anastomoses, placental location, and presence of artery-to-artery (AA) anastomoses were not significantly associated with IUFD of the donor twin.

Ultrasound Obstet Gynecol 2007; 30: 35–39.

Kontopoulos et al.

38 1.0 %AEDV > 30

Sensitivity

0.8

0.6

0.4

0.2

0.0 0.0

0.2

0.4 0.6 1 – Specificity

0.8

1.0

Figure 3 Receiver–operating characteristics curve to determine the optimal cut-off point of preoperative percent absent end-diastolic velocity (%AEDV) in the prediction of intrauterine fetal demise of the donor twin in patients with twin–twin transfusion syndrome treated with selective laser photocoagulation of communicating vessels.

DISCUSSION Our study suggests that %AEDV can be used as a prognostic factor in the prediction of IUFD of the donor twin in patients undergoing SLPCV for TTTS. This is a novel parameter that can be used in the assessment of patients with TTTS. Our data show that a preoperative %AEDV > 30 is associated with an increased risk of IUFD of the donor twin. The incidence of loss of the donor twin after laser surgery for TTTS has been reported in various ways. Ville et al.12 reported a 49% (22/45) incidence of IUFD of the donor, of which 86% (19) died within 24 h of surgery. In the more recently reported Eurofetus trial13 , 11.6% (16/138) of fetuses had died within a week of surgery, and IUFD occurred in 18.75% (27/144) of fetuses beyond 24 weeks’ gestation, with 50% (36/72) overall survival without neurological damage of donor twins at the age of 6 months. Hecher et al.14 reported a 63% survival of the donor (80/127) and a 72% survival of the recipient (92/127). Zikulnig et al.5 compared the survival rate of donor twins relative to the presence or absence of enddiastolic flow. In their study, AEDV was not associated with an increased risk of loss of the donor (13/23, 56% vs. 38/98, 38%, P = 0.12). We reported an incidence of 29.1% (32/110) of IUFD of the donor twin and 17.3% (19/110) of the recipient twin in a previous study7 . IUFD of the donor twin has been attributed, in part, to presumed placental insufficiency of the donor. Unfortunately, analysis of the placenta after IUFD to support this is hindered by fibrosis of the placental territory of the demised twin. In singletons, AEDV has been attributed to increased downstream placental

Copyright  2007 ISUOG. Published by John Wiley & Sons, Ltd.

resistance because of a reduction in the number of small muscular arteries in the tertiary stem villi, obliteration or abnormal development of the terminal villi15,16 . Other explanations, such as increased fetal blood viscosity17 or hypotension9,18 have also been invoked. AEDV has been associated with hypoxemia and acidosis that precede fetal heart rate changes19 , with a perinatal mortality rate in singleton fetuses with intrauterine growth restriction and AEDV of 40–70%20 . AEDV in TTTS has also been associated with poor perinatal outcome5,7,21,22 . Although the cause of AEDV in TTTS may be similar to that in singleton pregnancies, particularly because donor twins may have growth restriction and other signs of placental insufficiency, there is evidence to suggest that alternative mechanisms, such as the types of placental vascular anastomoses, may also operate7,8,18,23 . Indeed, our group has shown that AEDV can result from a decreased amount of individual placental territory (IPT), from the presence of AA anastomoses, or from both9 . Because IPT cannot be assessed reliably in patients with an IUFD, this study could not investigate the association between degree of %AEDV and IPT. However, we speculate that a higher %AEDV may be related to a smaller IPT9 . We also speculate that patients with IUFD of the donor twin may represent the extreme end of the spectrum in the relationship between %AEDV and decreased IPT. IUFD of the donor twin was associated with gestational age at the time of surgery as well as length of operating time. However, these factors are unlikely to be of clinical significance. Indeed, the median gestational age at procedure in patients with or without IUFD of the donor twin differed by only 1 week. Similarly, operating time differed by only 8 min. Thus, %AEDV, and particularly %AEDV > 30, appears to be the only clinically important parameter in predicting IUFD of the donor twin. Although presence or absence of end-diastolic flow and AA anastomoses were also significantly different between patients with and those without IUFD of the donor, logistic regression showed that these variables were surrogates of %AEDV. Our study is limited by its retrospective nature. This resulted in 55/127 (43%) patients not having digital images available for analysis. Nonetheless, the incidence of IUFD of the donor twin did not differ between the group of patients with AEDV in whom %AEDV was calculated and in the group in whom it was not. Thus, patients with calculated %AEDV probably represent a random sample of the AEDV population, from which statistical conclusions can be derived safely. In conclusion, increasing %AEDV is associated with an increased likelihood of IUFD of the donor twin in patients undergoing SLPCV for TTTS. A cut-off of 30% identifies approximately 75% of patients with AEDV destined to have an IUFD of the donor after SLPCV. Assessment of %AEDV during preoperative Doppler

Ultrasound Obstet Gynecol 2007; 30: 35–39.

Percent absent end-diastolic velocity in TTTS ultrasound examination may improve the counseling of TTTS patients undergoing evaluation for laser surgery.

ACKNOWLEDGMENTS We thank Mrs Karen Pomeroy and Mrs Lynette Kordos for their expertise in obtaining the sonographic images and Mr Michael Krueger, biostatistician, Wayne State University, for his invaluable assistance in the statistical analysis.

REFERENCES 1. Nicolini U, Nicolaidis P, Fisk NM, Vaughan JI, Fusi L, Gleeson R, Rodeck CH. Limited role of fetal blood sampling in prediction of outcome in intrauterine growth retardation. Lancet 1990; 336: 768–772. 2. Neilson JP, Alfirevic Z. Doppler ultrasound for fetal assessment in high risk pregnancies. Cochrane Database Syst Rev 2000; Issue 2: CD000073. 3. Zelop CM, Richardson DK, Heffner LJ. Outcomes of severely abnormal umbilical artery doppler velocimetry in structurally normal singleton fetuses. Obstet Gynecol 1996; 87: 434–438. 4. Quintero R, Morales W, Allen M, Bornick P, Johnson P, Krueger M. Staging of twin-twin transfusion syndrome. J Perinatol 1999; 19: 550–555. 5. Zikulnig L, Hecher K, Bregenzer T, Baz E, Hackeloer BJ. Prognostic factors in severe twin-twin transfusion syndrome treated by endoscopic laser surgery. Ultrasound Obstet Gynecol 1999; 14: 380–387. 6. Mari G, Detti L, Oz U, Abuhamad AZ. Long-term outcome in twin-twin transfusion syndrome treated with serial aggressive amnioreduction. Am J Obstet Gynecol 2000; 183: 211–217. 7. Martinez JM, Bermudez C, Becerra C, Lopez J, Morales WJ, Quintero RA. The role of Doppler studies in predicting individual intrauterine fetal demise after laser therapy for twintwin transfusion syndrome. Ultrasound Obstet Gynecol 2003; 22: 246–251. 8. Quintero RA, Martinez JM, Lopez J, Bermudez C, Becerra C, Morales W, Arroyo J. Individual placental territories after selective laser photocoagulation of communicating vessels in twin-twin transfusion syndrome. Am J Obstet Gynecol 2005; 192: 1112–1118. 9. Chang YL, Chmait RH, Bornick PW, Allen MH, Quintero RA. The role of laser surgery in dissecting the etiology of absent or reverse end-diastolic velocity in the umbilical artery of the donor twin in twin-twin transfusion syndrome. Am J Obstet Gynecol 2006; 195: 478–483. 10. Quintero R, Morales W, Mendoza G, Allen M, Kalter C, Giannina G, Angel J. Selective photocoagulation of placental vessels in twin-twin transfusion syndrome: Evolution

Copyright  2007 ISUOG. Published by John Wiley & Sons, Ltd.

39

11.

12.

13.

14.

15.

16.

17.

18.

19.

20.

21.

22.

23.

of a surgical technique. Obstet Gynecol Surv 1998; 53: s97–s103. Bermudez C, Becerra CH, Bornick PW, Allen MH, Arroyo J, Quintero RA. Placental types and twin-twin transfusion syndrome. Am J Obstet Gynecol 2002; 187: 489–494. Ville Y, Hyett J, Hecher K, Nicolaides K. Preliminary experience with endoscopic laser surgery for severe twintwin transfusion syndrome. N Engl J Med 1995; 332: 224–227. 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; 351: 136–144. Hecher K, Diehl W, Zikulnig L, Vetter M, Hackeloer BJ. Endoscopic laser coagulation of placental anastomoses in 200 pregnancies with severe mid-trimester twin-to-twin transfusion syndrome. Eur J Obstet Gynecol Reprod Biol 2000; 92: 135–139. Arduini D, Rizzo G. Doppler studies of deteriorating growthretarded fetuses. Curr Opin Obstet Gynecol 1993; 5: 195–203. Krebs C, Macara LM, Leiser R, Bowman AW, Greer IA, Kingdom JC. Intrauterine growth restriction with absent enddiastolic flow velocity in the umbilical artery is associated with maldevelopment of the placental terminal villous tree. Am J Obstet Gynecol 1996; 175: 1534–1542. Steel SA, Pearce JM, Nash G, Christopher B, Dormandy J, Bland JM. Correlation between Doppler flow velocity waveforms and cord blood viscosity. Br J Obstet Gynaecol 1989; 96: 1168–1172. Murakoshi T, Quintero RA, Bornick PW, Allen MH. In vivo endoscopic assessment of arterioarterial anastomoses: insight into their hemodynamic function. J Matern Fetal Neonatal Med 2003; 14: 247–255. Nicolaides KH, Bilardo CM, Soothill PW, Campbell S. Absence of end diastolic frequencies in umbilical artery: a sign of fetal hypoxia and acidosis. BMJ 1988; 297: 1026–1027. Karsdorp VH, van Vugt JM, van Geijn HP, Kostense PJ, Arduini D, Montenegro N, Todros T. Clinical significance of absent or reversed end diastolic velocity waveforms in umbilical artery. Lancet 1994; 344: 1664–1668. Mari G, Roberts A, Detti L, Kovanci E, Stefos T, BahadoSingh RO, Deter RL, Fisk NM. Perinatal morbidity and mortality rates in severe twin-twin transfusion syndrome: results of the International Amnioreduction Registry. Am J Obstet Gynecol 2001; 185: 708–715. Taylor MJ, Denbow ML, Duncan KR, Overton TG, Fisk NM. Antenatal factors at diagnosis that predict outcome in twintwin transfusion syndrome. Am J Obstet Gynecol 2000; 183: 1023–1028. Hecher K, Jauniaux E, Campbell S, Deane C, Nicolaides K. Artery-to-artery anastomosis in monochorionic twins. Am J Obstet Gynecol 1994; 171: 570–572.

Ultrasound Obstet Gynecol 2007; 30: 35–39.