The intrauterine fetal death (IUFD) ofone twin occurs most commonly during the first trimester. This phenomenon is known as a “vanishing twin” (Landy and Weingold, 1989). Although this may be associated with vaginal spotting, the loss of one conceptus is often not clinically recognized, and the prognosis for the surviving twin is excellent (Landy et al., 1986; Samuels, 1988; Prompeler et al., 1994).

Single IUFD in the second or third trimesters is much less common. Single IUFD is observed more often in association with monochorionic than with dichorionic placentation. When it occurs in a monochorionic gestation it may be associated with a worse outcome for the surviving co-twin.

The etiology of IUFD in a multiple pregnancy may be similar to singletons or unique to the twinning process. IUFD may be caused by genetic or anatomical anomalies, abruption, placental insufficiency, cord abnormalities such as a velamentous cord insertion, infection, and maternal disease including diabetes and hypertension (Ogunyemi et al., 1998; Collins, 2002; Craven and Ward, 2002; Simpson, 2002). In monochorionic pregnancies, IUFD may result from complications of the twin-to-twin transfusion syndrome (TTTS). Most often, the smaller, donor twin dies, but IUFD can occur in the larger recipient twin (D’Alton and Simpson, 1995). In addition, monoamniotic twins are at increased risk of cord entanglement and subsequent IUFD (Colburn and Pasquale, 1982). Similar to singletons, the etiology of IUFD often remains elusive (Santema et al., 1995).

Single IUFD in a multiple gestation can adversely affect the surviving fetus or fetuses in two ways: (1) risk for multicystic encephalomalacia and multiorgan damage in monochorionic pregnancies and (2) preterm labor and delivery in both dichorionic and monochorionic twins resulting in prematurity.

Multi cystic encephalomalacia (cystic lesions in the cerebral white matter distributed in areas supplied by the anterior and middle cerebral arteries) is associated with profound neurologic handicap. The risk of multicystic encephalomalacia following single IUFD in a monochorionic pregnancy may be as high as 20% for the surviving co-twin (Melnick, 1977; Wessel and Schmidt-Gollwitzer, 1988; Yoshida and Matayoshi, 1990; Dudley and D’Alton, 1986; Eglowstein and D’Alton, 1993; D’Alton and Simpson, 1995; Pharoah and Cooke, 1997; Pharoah and Adi, 2000).

Two theories have been suggested to explain the pathophysiology of multiorgan damage in surviving monochorionic twins. One theory is based on the premise that the retained demised fetus produces thromboplastic materials that traverse the anastomoses between the placentas resulting in disseminated intravascular coagulation. This would lead to infarction and cystic change in numerous organ systems including the kidneys, lungs, spleen, liver, and brain (Bulla et al., 1987). Although several autopsy reports have supported this hypothesis (Moore et al., 1969; Yoshioka et al., 1979; Szymonowicz et al., 1986), this mechanism of injury is no longer thought to be plausible.

The current widely accepted theory suggests that blood from the surviving twin may rapidly “back-bleed” into the demised twin through placental anastomoses, a form of acute fetofetal transfusion (Fusi et al., 1991). The demised twin may be come congested while the surviving twin may become anemic. If the hypotension is significant, the surviving twin is at risk for ischemic damage to vital organs (D’Alton et al., 1984; Fusi et al., 1991; Okamura et al., 1994).

In 1991, Fusi et al. reported a case where cerebral and renal lesions were pathologically confirmed in a deceased neonate who had experienced IUFD of a co-twin. Autopsy demonstrated acute blood transfusion from the surviving twin to the IUFD. Hematocrit at delivery was consistent with anemia (Fusi et al., 1991). Fusi’s case suggests the need for intervention before IUFD in monochorionic twins to avoid permanent multiorgan damage. Nicolini et al. further supported this concept when they described eight monochorionic pregnancies complicated by a single IUFD (Nicolini et al., 1998). Five of the eight pregnancies underwent blood sampling prior to IUFD. Four of these five who had died were not anemic prior to death and neither were their co-twins. All survivors sampled within 24 hours following the death of their co-twin were anemic.

Since the injury seems to occur at the time of IUFD, immediate delivery of the co-twin following single IUFD in a monochorionic pregnancy does not seem to improve out-comes but rather adds to the risk of prematurity. Perfectly timed intravascular transfusion of blood or fluid immediately prior to the death of the moribund twin has been suggested as a therapeutic intervention to improve outcomes (Fusi et al., 1991; Nicolini et al., 1998; Nicolini and Poblete, 1999). In our opinion, this is unproven and veryunlikelyto have any significant clinical role.

It is unclear at which gestational age IUFD in a monochorionic pregnancy may result in adverse sequelae for the surviving co-twin. It is often taught that the risk for multicystic encephalomalacia begins in the late second and third trimesters (D’Alton and Simpson, 1995). Neuropathology has not been able to pinpoint the exact timing of insult (D’Alton and Dudley, 1989). According to Pharoah et al., it is possi-ble that early demise (“vanishing twin”) in a monochorionic pregnancy secondary to acute and early TTTS could be an explanation for a portion of cases of unexplained cerebral palsy (Pharoah and Cooke, 1997; Pharoah and Adi, 2000). Until further research is completed, this theory can neither be supported nor refuted (Blickstein, 1998).

Of note, we recently reported a case of an IUFD ofone twin at approximately 13 weeks in a monochorionic gestation that resulted in multicystic encephalomalacia in the surviving twin. Ultrasound and Magnetic Resonance Imaging (MRI) were performed at approximately 21 weeks demonstrating the neuroradiological pathology (Figures 118-1 and 118-2). The patient was counseled regarding the poor prognosis and opted for termination. The multicystic encephalomalacia was confirmed by pathology, although the exact timing of the injury could not be determined (Weiss et al., 2004).

In addition, studies have demonstrated that IUFD of one twin can result in preterm delivery in both monochorionic and dichorionic pregnancies (Carlson and Towers, 1989; Peterson and Nyholm, 1999). In a study by Carlson and Towers of 17 twin pregnancies complicated by IUFD, 76% of these pregnancies were delivered before 37 weeks. Eighty-six percent of the patients delivering prematurely presented in active labor (Carlson and Towers, 1989). This is almost double the background incidence of reported preterm delivery in twins.

Besides fetal risks, there is a theoretical possibility for maternal consumptive coagulopathy in twin pregnancies complicated by single IUFD (Anderson et al., 1990). It has been established that retention of a dead fetus for 4 to 5 weeks in a singleton pregnancy results in an increased risk of maternal consumptive coagulopathy (Pritchard and Ratnoff, 1955; D’Alton and Simpson, 1995). It was originally estimated that there was a 25% incidence of maternal disseminated intravascular coagulation when fetal death occurred in multiple gestations (Landy and Weingold, 1989). However, only a few cases of laboratory evidence rather than clinical evidence of coagulopathy have been reported under these circumstances, and the 25% incidence is likely an overestimation (Anderson et al., 1990; Carlson and Towers, 1989; D’Alton and Simpson, 1995). In one series of 16 pregnancies complicated by IUFD of one twin, no cases of maternal disseminated intravascular coagulation were found (Fusi and Gordon, 1990). In another series, transient fibrin-split products and hypofibrinogenemia were found in 2 of 20 cases of single IUFD in twins, neither of which was clinically apparent or required medical therapy (Eglowstein and D’Alton, 1993). It is also reassuring to note that no cases of clinically significant coagulopathy have been reported in the extensive literature on selective termination and multifetal pregnancy reduction.

Ultrasound during the first trimester has demonstrated that the incidence of twins is greater than previously presumed (Landy and Weingold, 1989). Single IUFD in the first trimester may occur in up to 50% of cases of twin pregnancies (Varma, 1979).

Single IUFD of one fetus in a multiple gestation in the second or third trimesters is much less common complicating approximately 0.5% to 6.8% of twin pregnancies (Benirschke, 1961; Litschgi and Stucki, 1980; D’Alton et al., 1984; Hanna and Hill, 1984; Enbom, 1985; Carlson and Towers, 1989; Fusi and Gordon, 1990; Kilby et al., 1994; Johnson and Zhang, 2002). Monochorionic twins are at increased risk for single IUFD compared to dichorionic twins. It is estimated that there is a threefold to fourfold increase in intrauterine death with monochorionic twins as compared with dichorionic twins (Burke, 1990; Kilby et al., 1994). It is interesting to note that because of increasing use and success of assisted reproductive technologies, the incidence of monochorionic twins is increasing and may be greater than 10 times the spontaneous rate (Blickstein, 2005).

In addition, a single fetal death is also more common among twins with a structural abnormality (Kilby et al., 1994). Demise of both fetuses in a twin pregnancy hasbeen reported infrequently (Rydhstroem, 1996). In high-order multiples, however, demise of a single fetus may be more common. Studies have indicated single IUFD rates ranging from 4.3% to 17% in triplet pregnancies (Gonen et al., 1990; Borlum, 1991; Johnson and Zhang, 2002).