Outcome of isolated congenital complete heart block diagnosed in utero

Transcription

1 19 Departments of Fetal and Paediatri Cardiology, Guy's Hospital, London A M M Groves L D Allan E Rosenthal Correspondene to: Dr A Groves, 15th Floor, Guy's Tower, St Thomas's Street, London SE1 9RT. Aepted for publiation 15 August 1995 Heart 1996;75: Outome of isolated ongenital omplete heart blok diagnosed in utero Alison M M Groves, Lindsay D Allan, Eri Rosenthal Abstrat Objetive-To establish identifiable prenatal fators in fetal heart blok whih might predit death in utero, the need for intervention, or the probability of paemaker requirement. Setting-Tertiary referral unit for fetal ehoardiography. Subjets-36 fetuses with ongenital omplete heart blok and struturally normal hearts identified between 198 and Methods-Maternal anti-ro antibody status was doumented. Prenatal variables examined inluded absolute heart (ventriular) rate, hange in rate, and development of hydrops fetalis. Postnatally, heart rate, need for paing, and the indiations for paing were detailed. Results-Of the total of 36 patients, there are 24 survivors; 11 are paed. Of those fetuses whih died, two were eletively aborted for severe hydrops, seven died in utero, two were immediate postnatal deaths, and one was an unrelated infant death. The trend was for the heart rate to derease during fetal life and postnatally. Fetuses with deteriorating ardia funtion did not always show the lowest heart rates. Bradyardia of less than 55 beats/min in early pregnany or rapid derease in heart rate prenatally were poor prognosti signs. Hydrops was also assoiated with bad outome, 1 out of the 12 hydropi fetuses dying (83%). Of 1 fetuses presenting with a heart rate above 6/min, nine survived of whom three required paing. Of seven presenting with heart rates of S/min or less, only three survived and two of these required paing. Of the two fetuses with negative maternal anti-ro antibody status one died in utero and one required heart transplantation after paemaker insertion. Conlusions-Isolated omplete heart blok identified in fetal life does not always have a good prognosis. An individual heart rate does not aurately predit the outome in utero or the need for postnatal paing. Regular, areful monitoring during pregnany is required in order to optimise are and timing of any interventions. (Heart 1996;75:19-194) Keywords: fetal heart blok; paing; hydrops fetalis; ardia funtion Morquio, in 191, was first to desribe "omplete heart blok" in several members of one family.' The first eletroardiographi analysis was reported in Identifiation in the fetus ame in 1945, when the diagnosis was made by ausultation and eletroardiography.3 Eletroardiography is tehnially diffiult and ehoardiography has beome the method of hoie for the detetion and analysis of fetal arrhythmias.46 An assoiation between fetal isolated omplete heart blok and maternal onnetive tissue disease has been desribed.78 Pathologial hanges have been diretly related to irulating maternal anti-ro antibodies, even in the absene of linial disease.9 Histologially there is fibrous replaement of the onduting tissue.',1 In 1988 we reported our linial experiene of omplete heart blok in the fetus: 16 ases with isolated omplete heart blok and 21 of heart blok with strutural heart disease (17 with left atrial isomerism). Fetal outome with isolated heart blok was good, ompared with a mortality of 85% in those with strutural heart disease.'2 It has sine beome evident that ardia failure and death an our in isolated heart blok.'"'5 This natural history has led to interventions inluding premature delivery,'6 administration of inotropes,'7 steroids and/or plasmapheresis'8 and transabdominal paing of the fetus.'9 Sine 198, we have diagnosed isolated omplete heart blok in 36 fetuses, inluding one set of twins. An analysis of the outome of the group-the largest single entre series to date-was undertaken in order to establish any risk fators identifiable prenatally whih might predit in utero death or paemaker requirement. These would be relevant for prenatal intervention, site of delivery, timing and management of labour, and parental ounselling. Methods Between 198 and January 1994, 36 fetuses with isolated ongenital omplete heart blok were identified from 1 6 referrals for fetal ehoardiography. The reason for referral was detetion of a fetal bradyardia in 34 patients and presene of maternal antibodies in threeone with a previous hild with isolated heart blok. Antibody status was established in 33 patients using enzyme linked immunosorbant assay (ELISA). One patient was on high dose steroids for systemi lupus erythematosus before oneption. The diagnosis was established on ross

2 Outome of isolated ongenital omplete heart blok diagnosed in utero 1 9 = 8 CD C 7._m E a) 6 Cn Co a) m 5 4 t 3 &L 2 1 Table 1 Outome of 36fetuses (one set twins) Antibody status-34 tested Ro positive Clinially apparent onnetive tissue disease Previous sibling with doumented CHB Presene of hydrops Early Late Periardial effusion only Outome Termination of pregnany (TOP) Intrauterine death (IUD) Neonatal death (NND) Infant death (INFD) Alive and well, not paed Alive, paed 32 (94%) 7 (2%) 3 (8 5%) (1 temporarily paed) 1 (non-ardia) (1 transplant, 1 alive with neurologial problems) setional imaging using an Advaned Tehnial Laboratories mark 4 or a Hewlett Pakard 772A ultrasound sanner. At eah visit heart rate was alulated by the average of three M mode or Doppler readings, preferably at rest, and ardia funtion evaluated using M mode shortening fration and arterial Doppler peak veloities. Evidene of intrauterine ardia failure-skin oedema, asites, and periardial or pleural effusion-was reorded. With inreasing experiene patients were # n = 36 (23) (16) (14) (8) (5) (2) (1) Weeks after estimated time of oneption Years after birth Figure 1 Kaplan-Meyer survival urve for ongenital omplete heart blok deteted prenatally. This urve shows that the major attrition ofpatients in the prenatal series is antenatal. The non-ardia death has been exluded. There was no fetal loss before 27 weeks gestation. Numbers in parenthesis are the numbers ofpatients alive at the follow up interval. Table 2 95 % onfidene interval Survival proportion. 95 % onfidene interval Indiations for paing Asymptomati Symptomati Bradyardia Exerise + long QT Bradyardia CCF Synope FTT intolerane + pauses + exit blok No of ases Age at 4d, 14d 9m, 9m 4m 8y 8d, 12d, 4m 4y, 6y paing CCF, ongestive ardia failure; FTT, failure to thrive....*,o reviewed more frequently, four to six weekly initially, but two weekly if any signs of ardia failure were noted. Postnatal eletroardiography onfirmed the diagnosis in all live deliveries. The heart rates were reorded by linial observation or by 12-lead eletroardiogram and by ambulatory eletroardiographi analysis. The age and riteria used for paing were reorded. Details in all surviving patients were obtained from the relevant paediatri ardiology entre. Neropsy analysis of the ardia struture and the histology of the onduting tissue were sought where possible. The results were assessed for signifiane using the %2 test and signifiane was assumed atp < 5. Results PRENATAL COURSE The findings in the total group of 36 ases of prenatal isolated omplete heart blok are summarised in table 1. Of nine in utero deaths, only one fetus was not hydropi and had no signs of deteriorating ardia funtion. Neropsy examination did not reveal any abnormalities other than fibrosis of the bundle. One of the in utero fetal deaths had negative maternal anti-ro antibody. Histology showed interruption of the bundle of His in ontrast to the fibrosis of those with antibody. HYDROPIC FETUSES Twelve patients developed signifiant intrauterine hydrops. Two of these survived-one showed spontaneous resolution and the other had a ourse of sympathomimeti treatment whih appeared to reverse the deterioration in ventriular funtion, with resolution of hydrops. Neither has required paing postnatally. Three mothers were given steroids without any improvement in the fetus. Two pregnanies were eletively interrupted in view of the severity of fetal hydrops. Two of the hydropi infants were delivered prematurely. One died of ardia and renal failure before paing was instigated. The other died, despite C'Z. Co ) ~ CL 9 r 8 F _ 15 Figure 2 + * * + + e I* Non-Paed *Paed + Died Gestation (weeks) The relation ofpresenting heart rate to outome. There is a wide satter overaul with some lustering of the patients with poor outome at lower rates. 191

3 192 Groves, Allan, Rosenthal Figure 3 Prenatal heart rates offetuses sanned more than one, grouped aording to outome. Eah set of symbols joined by a line represents a patient. Figure 4 Averaged heart rates at presentation, birth, and most reentfollow up ofpatients who died in utero or early neonatal period (exluding the pertussis death), those who survived having been paed, and survivors who have not needed paing. The differenes were not statistially signifiant. IUD, intrauterine death; NND, early neonatal death. Table 3 a) I) I 8r 7K 6K Patients who died (n = 6) 5K 8 7 H 6K 5 H 4 H 4 H 3n ou a I sot' Non-paed, n =13 * Paed, n =11 - Died, n =11 'ZSt' paing, at 1 days. Four fetuses developed an isolated periardial effusion without evidene of ventriular dysfuntion. These resolved spontaneously postnatally. POSTNATAL COURSE One patient died in infany of pertussis infetion. To date there has been no further mortality (fig 1). One patient who had a satisfatory heart rate on 24 h monitoring had a ardia arrest at 4 months and was paed after resusitation. She now has a severe neurologial handiap. Retrospetive analysis showed a prolonged QT interval of 648 ms and heart rate 6/min with no ventriular arrhythmia noted. One of the survivors had negative maternal anti-ro status. This patient was paed but required ardia transplantation for ardiomyopathy. The exised heart speimen showed interruption of the bundle of His. Outome ofprenatally diagnosed isolated heart blok Heart rate at Died Died Paed Paed Not presentation Total Hydropi (IUD) (PND) < lyr > l yr paed > it * < : *Inludes two pregnanies terminated for severe hydrops. IUD, intrauterine death; PND, postnatal death, inludes one (t) non-ardia and two (t) delivered prematurely. Patients requiring paing (n = 1) 8 Il Gestation (weeks) 7 _- 6 5K 4 H Patients not requiring - paing (n = 1) Of the 24 surviving patients with isolated omplete heart blok diagnosed in utero, 11 (46%) have required paing so far. Maximum follow up duration is 12 years, minimum six months, mean 5j years. Table 2 outlines the age and indiation for paing. IN UTERO HEART RATE One patient had a normal heart rate at 18 weeks gestation whih evolved into omplete heart blok by 22 weeks. Otherwise, heart rate at presentation varied between 45 and 8 beats/min, with overall a higher rate earlier in gestation. Fetal heart rate inreased with ativity (fig 2). Most fetuses showed no signifiant hange in heart rate with advaning gestation, but of 25 who had more than one examination, nine showed a derease of more than 5 beats/min. Of these, five died in utero (P <.1), two showed an inreased rate, and one of these fetuses subsequently died (fig 3). Heart rate at presentation was 51 (SD 6 8) beats/min for those who died, 53 (8-1) beats/min in those who survived but required paing, and 64 (8 3) beats/min in those survivors not requiring paing (fig 4). However, the differene between the heart rate at presentation of infants paed but alive and those who died was not signifiant, and there was no signifiant differene between rates at presentation of the paed versus non-paed survivors. Of 1 fetuses presenting with a heart rate above 6/min, nine survived (9%) and only three have required paing so far (33% of survivors). Meanwhile, of seven presenting with a rate of 5 or less, only three survived (42%) and two of these three have required paing (table 3). A presenting heart rate less than 55/min before 28 weeks gestation was assoiated with death in five and permanent paemaker in the other one of six fetuses. This reahed statistial signifiane on X2 analysis (P < -1; fig 2). Disussion Isolated ongenital omplete heart blok in the fetus has been well desribed. The assoiation with maternal anti-ro antibody has also been N.

4 Outome of isolated ongenital omplete heart blok diagnosed in utero widely doumented. Intrauterine ardia failure leading to fetal or neonatal loss has beome more widely reported.' In our urrent series the greatest attrition ourred prenatally, heart failure affeting 12 of our 36 ases. Perinatal loss ourred in 1 of these ases, whih is omparable to hydropi deaths in all four of 26 ases in a ombined report from three entres.'5 We have not shown an absolute relation between the initial ventriular rate and outome but agree with other investigators that rates below 55/min during early pregnany have a worse prognosis.'3 15 However the wide satter makes it diffiult to give a prognosis in any one ase (fig 2). Our findings are omparable with the postnatal series without strutural ardia defets reported by Sholler and Walsh.2 The majority (56%) of fetuses maintained the same rate. In 36% heart rate fell with advaning gestation, and where it dereased by 5 beats/min or more the mortality was 55%. Two patients in the group showed a small inrease in heart rate, indiating some variability with ativity already doumented in hildren with heart blok.2' One of these fetuses died in utero, whih was perhaps surprising as the ability to inrease heart rate with exerise is assoiated with favourable outome postnatally Suessful maternal prophylaxis has been laimed for steroids and plasmapheresis in individual ases. However, as the reurrene risk is only 8-2%, treatment effiay has not been established.9 Prenatal intervention in the hydropi affeted fetus has also been tried. Three suh patients were given steroids with no suess, but we have had suess with salbutamol in one patient in our unit.'7 In one report diret intrauterine paing was tehnially suessful for some hours before fetal death.'9 Long term problems would be potential sepsis, displaement of the wire, or fetal limb damage from it. Future devies might be implanted under the fetal skin. The major attrition in our series is prenatally and immediately postnatally in those delivered prematurely for worsening hydrops. All but one was hydropi.'3 There has been no mortality related to heart blok beyond the neonatal period. The dereasing heart rate seen in serial prenatal sans often ontinued into postnatal life, perhaps beause of fibrosis extending to the bundle of His." Endoardial paing in hildhood is now tehnially reliable although not undertaken without lear indiations. Permanent paing was required in 11 patients, 55% of whom were symptomati. Asymptomati ases did not have absolute indiations for paing, but low heart rates and either long QT or exit blok with pauses were the basis of the deision to pae in our unit and other paediatri ardiology entres. The ombination of a low heart rate with long QT interval had a signifiantly poor outome in a postnatal multientre series. The patient who was paed only after ardia arrest illustrates the importane of assessment of QT interval.25 The two ases whih showed repeated negative maternal anti-ro antibody both had poor outomes. One died prenatally and the other required ardia transplantation-both hearts showing interruption of the bundle of His of a familial type. This may have represented a new mutation or inomplete family history. CONCLUSIONS Reviewing our series of patients with isolated omplete heart blok we found a heart rate below 55 beats/min in early pregnany, or one whih drops below 5 beats/min, is assoiated with greater likelihood of poor outome. The likelihood of requiring early paemaker insertion, however, annot be predited with ertainty from the prenatal heart rate. These patients should therefore be monitored at frequent intervals, every one to two weeks, or monthly for those at lower risk, searhing partiularly for early evidene of ardia deompensation. In this way the optimum deision would be made with respet to possible antenatal treatment, early delivery, perinatal monitoring without ardiotoographi assistane, and management of the hydropi neonate. We would like to aknowledge the assistane of Dr J Till, Royal Brompton and National Heart Hospital, for outome details. 1 Morquio L. Sur un maladie infantile araterisee par des modifiations permanente du pouls, des attaques synopales et epileptiformes et la mort subite. Arh Med Enfants 191;4: Yater WM. Congenital heart blok: review of the literature report of a ase with inomplete heterotaxy: the eletroardiogram in dextroardia. Am Jf Dis Child 1929;38: Plant RK, Steven RA. Complete atrioventriular blok in a fetus. Am Heart_J 1945;12: Crawford D, Chapman M, Allan LD. The assessment of persistent bradyardia in prenatal life. Br J Obstet Gynaeol 1985;92: Allan LD, Anderson RH, Sullivan ID, Campbell S, Holt DW, Tynan M. Evaluation of fetal arrhythmias by ehoardiography. Br HeartJ_ 1983;5: Gembruh U, Hansmann M, Redel DA, Bald R, Knopfle G. Fetal omplete heart blok: antenatal diagnosis, signifiane and management. Eur J Obst Gyneol 1989;31: Hull D, Binns BAO, Joye D. Congenital heart blok and widespread fibrosis due to maternal lupus erythematosus. Arh Dis Child 1966;41: MCue CM, Mantakas ME, Tingelstad JB, Ruddy S. Congenital heart blok in newborns of mothers with onnetive tissue disease. Cirulation 1977;56: Sott JS, Maddison PJ, Taylor PV, Essher E, Sott, Skinner RP. Connetive tissue disease, antibodies to ribonuleoprotein, and ongenital heart blok. N Engl J Med 1985;312: Lev M, Silverman J, Fitzmaurie FM, Paul MH, Cassels DE, Miller RA. Lak of onnetion between the atria and the more peripheral ondution system in ongenital atrioventriular blok. Am J Cardiol 1971 ;27: Ho SY, Fagg N, Anderson RH, Cook A, Allan LD. Disposition of the atrioventriular ondution tissues in the heart with isomerism of the atrial appendages: its relation to ongenital omplete heart blok. J Am Coll Cardiol 1992;2: Mahado MVL, Tynan MJ, Curry PVL, Allan LD. Fetal omplete heart blok. Br Heartj7 1988;6: Kleinman CS, Donnerstein RL. Ultrasound assessment of ardia funtion in the intat human fetus. JT Am Coll Cardiol 1985;5:84-94S. 14 Serwer GA, Vermillion RP, Snider AR, Dik M, Crowley DC, Hayashi RH. Prognosti indiators for fetuses with in utero diagnosed omplete heart blok [abstr]. Cirulation 1988;78(suppl II):II Shmidt KG, Ulmer HE, Silverman NH, Kleinman CS, Copel JA. Perinatal outome of fetal omplete atrioventriular blok: a multienter experiene.3am Coll Cardiol 1991;17: Martin TC, Arias F, Oglander DS, Hoffman RJ, Marburger JP, Maurer MM. Suessful management of ongenital atrioventriular blok assoiated with hydrops fetalis. J Pediatr 1988;112: Groves AMM, Allan LD, Rosenthal E. Therapeuti use of inotropes in omplete heart blok in the fetus [abstr]. Br 1993;69:17s. Heartr. 193

5 194 Groves, Allan, Rosenthal ABSTRACTS IN CARDIOLOGY 18 Barlay CS, Frenh MAH, Ross LD, Sokol RJ. Suessful pregnany following steroid therapy and plasma exhange in a woman with anti-ro (SS-A) antibodies. Case report. Br J Obstet Gynaeol 1987;94: Carpenter RJ, Strasburger JF, Garson A, Smith RT, Deter RL, Engelhardt HT. Fetal ventriular paing for hydrops seondary to omplete atrioventriular blok. J Am Coll Cardiol 1986;8: Sholler GF, Walsh EP. Congenital omplete heart blok in patients without anatomi ardia defets. Am Heart J 1989;118: Mihaelsson M, Engle MA. Congenital omplete heart blok: an international study of the natural history. Cardiovas Clin 1972;4: Rheumati fever: pathogenesis revealed 22 Dewey RC, Capeless MA, Levy AM. Use of ambulatory eletroardiographi monitoring to identify high-risk patients with ongenital omplete heart blok. N Engl J Med 1987;14: Pinskey WM, Gillette PC, Garson A, MNamara DG. Diagnosis, management, and long-term results of patients with ongenital omplete atrioventriular blok. Pediatris 1982;69: Karpawih PP, Gillette PC, Garson A, Hesslein PS, Porter C-B, Manamara DG. Congenital omplete atrioventriular blok: linial and eletrophysiologi preditors of need for paemaker insertion. Am J Cardiol 1981;48: Essher E, Mihaelsson M. Q-T interval in ongenital omplete heart blok. Pediatr Cardiol 1983;4: Generations of medial students have been inonsist -enies. The Ashoff body has no hisfeatures that suggest humoral dam- taught onepts about the pathogenesis of tologial rheumati fever whih have great appeal. The age and tbound gamma globulin is not present onept of "moleular mimiry" lays down in the tisesues. The present study shows that T that ertain omponents of the wall of ertain ells fronn the hearts of subjets who have had strains of steptooi fortuitously show lose rheumatii fever an reognise both ardia moleular similarity with onnetive tissue and bat4 erial wall antigens and the damage to omponents of heart valves and myoardium the heartt is probably ell based-a small but and that antibodies produed against the ba- importanit hange in thinking terial antigen ross reat against the heart, ausing damage. The theory always had M J DAVIES Heart: first published as /hrt on 1 February Downloaded from on 9 April 219 by guest. Proteted by opyright.