Hemodynamic Effects of Chronic Prenatal Ventricular Pacing for the Treatment of Complete Atrioventricular Block
Background Increasing the heart rate of the fetus with cardiac failure caused by complete AV block (CAVB) may allow delivery of a full-term, stable neonate with preserved ventricular function. Direct fetal pacing may be a feasible method to achieve this, but the effect of pacing on the structure and function of the rapidly developing fetal heart is unknown.
Methods and Results CAVB was created in fetal lambs at 80% gestation by cryoablating the AV node. Epicardial ventricular pacing at 130 bpm was achieved by use of a pacemaker placed under the pectoral muscles. The fetus was returned to the uterus and allowed to continue to term. Ventricular function was assessed 1 week after birth in 7 lambs with CAVB and 10 control lambs. By use of the conductance catheter technique, the end-systolic pressure-volume relationship was determined at different heart rates, pacing conditions, and inotropic states. The contractility was not different between the two groups at their baseline heart rates and rhythms or when they were paced synchronously compared with asynchronously. Also, both groups responded significantly and similarly to inotropic manipulation, indicating preserved contractile reserve. Finally, in both groups, increased heart rates were associated with increased contractility, indicating an intact force-frequency relationship.
Conclusions We conclude that chronic epicardial ventricular pacing is well tolerated by the fetus, can be successfully applied as a treatment for CAVB, and does not adversely affect myocardial function in the rapidly developing, immature heart.
Congenital complete AV block (CAVB) remains an uncommon but problematic fetal condition. Hydrops fetalis occurs in ≈50% of fetuses with CAVB.1 In this setting, CAVB is almost uniformly fatal, reflecting the inadequacies of current therapeutic options and the often-associated myocardial dysfunction and congenital heart defects. With the advent of newer echocardiographic and Doppler ultrasound techniques, various fetal arrhythmias, including CAVB, are now diagnosed prenatally more often.2 3 4 5
Given the high mortality of CAVB, novel treatment modalities must be considered. One proposed new therapy is prenatal epicardial ventricular pacing. However, before application, it is important that we establish the morphological and functional consequences of epicardial ventricular pacing on the immature, rapidly developing fetal heart.
In this article, we report our findings of chronic epicardial ventricular pacing in a fetal model of CAVB. We created CAVB in fetal lambs by cryoablating the AV node. The fetuses subsequently underwent ventricular pacing to term. After birth, the left ventricular (LV) function was assessed by use of the conductance catheter method, and the hearts were submitted to pathology for evaluation.
After sedation (ketamine 10 mg/kg IM) and induction of spinal anesthesia (tetracaine 20 mg), 34 time-dated ewes at 115 to 120 days’ gestation (term, 145 days) underwent surgery. A percutaneous external jugular line was placed for intravascular infusion of fluids and medications. A midline laparotomy was performed to expose the uterus. The position of the fetus was identified, and a hysterotomy was performed over its right thorax. Anesthesia was induced either by direct injection of tetracaine 3 mg directly into the fourth ventricle of the brain or with ketamine (10 mg/kg IM), and paralysis was brought about with succinylcholine (20 mg IM). A right thoracotomy was performed in the fifth intercostal space, and a submuscular pocket was created over the right thorax. The right middle and lower lobes of the lung were carefully retracted to expose the inferior vena caval (IVC) junction with the right atrium. An oblique pericardiotomy was performed, and pericardial stay sutures were placed to improve exposure of the coronary sinus. An epicardial pacing lead (Medtronic 6500 Temporary Pacing Lead, Medtronics) was sutured onto the right ventricle and connected to an implantable pacemaker set in VVI mode at 130 ppm (Minix, Medtronics). This pacemaker is of adequate dimensions to permit placement of the generator within a subcutaneous pocket. Next, a straight, 3.5-mm cryoablation probe (Frigitronics Super Cool Glaucoma Probe 164, Frigitronics ECE-82 cryoablation unit, CooperVision Co) was placed on the external surface of the heart at the right atrial–coronary sinus juncture over the nodal triangle for 5 minutes at −80°C to cryoablate the AV nodal conduction tissue. During the ablation, the ventricle was intermittently paced for 10 beats every 30 seconds. The ablation was repeated as necessary until no ventricular escape rhythm was noted. This technique of creating AV block in the fetus was originally described by Assad and associates.6 After CAVB was attained, ventricular pacing was initiated. The pacemaker generator was placed in the subcutaneous pocket. The fetal incision was closed, and the fetus was returned to the uterus. Amniotic fluid lost during surgery was replaced with warm, normal saline. Antibiotics (penicillin 1×106 U and gentamicin 100 mg) were added to the amniotic fluid before closure of the uterus. The ewe recovered from anesthesia and was allowed to progress to full-term gestation and normal delivery.
Study Protocol on Neonatal Lambs
After birth, the lambs were closely observed to assess their perinatal status. At 1 week of life, 7 lambs with CAVB and 10 control lambs were studied according to a protocol approved by the Committee for Animal Research at the University of California, San Francisco. Each lamb was premedicated with ketamine hydrochloride (5 to 10 mg/kg IM) for placement of a peripheral arterial and venous line in the leg. An intravenous infusion of lactated Ringer’s solution was maintained throughout the study at 4 to 8 mL·kg−1·h−1. The lamb was anesthetized (ketamine hydrochloride 5 to 10 mg·kg−1·h−1), paralyzed (pavulon hydrochloride 0.2 mg/kg), and intubated. The lamb was mechanically ventilated to maintain its arterial Po2 and Pco2 in normal physiological ranges throughout the duration of the study. ECG leads were placed to monitor heart rate and rhythm.
A midline sternotomy and pericardotomy were performed to expose the heart and great vessels. With a 5F pressure-tip catheter (Millar Instruments, Houston, Tex), pressure tracings were obtained from the right atrium, left atrium, and pulmonary artery and digitized on-line to a Macintosh IIx computer (Apple). A pulmonary artery infusion line was then placed for administration of study drugs.
The Millar pressure-tip catheter was placed into the LV cavity. Its position was verified through pressure tracings on an oscilloscope (Digital Recording Oscilloscope, model 1604, Gould). A 7F sheath was then placed into the ascending aorta through which a 6F 10-electrode conductance catheter (Webster Laboratories) was guided so that at least one electrode on the catheter lay above the aortic valve and the most distal electrode lay in the apex of the left ventricle. The position of the conductance catheter was verified echocardiographically. Individual volume segment signals were examined during ventricular systole and diastole, and the segments that moved synchronously were combined to represent that ventricular volume during subsequent data analysis. Total volume was also sent to the X-Y oscilloscope with pressure and total volume displayed either as separate signals versus time or as pressure-volume loops throughout the study. Finally, a 5F atrial septostomy balloon catheter (American Edwards Laboratories) was introduced through the right atrial appendage into the IVC for IVC occlusion. Its position was assessed by palpation and confirmed during balloon inflation by the progressive decrease in LV volume on the oscilloscope.
In the chronically paced lambs, temporary ventricular and atrial pacing leads were sutured to the epicardium. Grounding leads were sutured to the skin. The ventricular lead and one of the ground leads were connected to an external pacemaker (model 5328 Programmable Stimulator, Medtronics). The implanted pacemaker was then dissected out of the subcutaneous pocket, its pacing lead was transected, and external pacing was initiated at 130 bpm. Persistence of CAVB was confirmed by the absence of antegrade conduction during a trial of atrial pacing at 200 bpm. When the protocol required AV sequential pacing, the leads were connected to an AV sequential pacemaker generator (model 5330 Programmable Stimulator, Medtronics). In the control lambs, right atrial and ventricular pacing leads and grounds (6500 Temporary Pacing Lead, Medtronics) were placed and connected to an external stimulator for atrial or ventricular pacing as determined by the study protocol.
Data Acquisition and Analysis
The pressure-tip catheter was connected through a conditioning amplifier (model 5900 Signal Conditioner, Gould) to a 12-bit analog-to-digital convertor and to an X-Y oscilloscope (model 1604 Digital Recording Oscilloscope, Gould). The conductance catheter was connected to a Sigma five-signal conditioner processor (Leycom) that supplied the current, computed instantaneous and continuous segmental conductances, and generated analog output to the oscilloscope for generation of pressure-volume loops and to the analog-to-digital convertor. The principles behind conversion of segmental conductances to absolute volumes have been discussed extensively by previous authors.7 8 9 10 11 We did not correct measured volumes to absolute volumes by subtracting parallel conductance of contiguous structures because we were interested only in the slope of the end-systolic pressure-volume relationship as our index of contractility (see below). Pressure and volume data input to the analog-to-digital convertor was digitized at 200 Hz and acquired and recorded on the hard disk of a Macintosh IIx computer with software designed at the University of California, San Francisco, in the LabView programming language (National Instruments).
In each lamb, data were acquired continuously for 20 seconds: during 5 seconds of a stable hemodynamic state and then during gradual IVC occlusion by inflation of the atrial septostomy balloon with 2.0 cm3 NaCl solution. This was performed while the ventilator was disconnected under three inotropic conditions: (1) at baseline, (2) during infusion of 10 μg·kg−1·min−1 dobutamine, and (3) during infusion of 150 to 200 μg·kg−1·min−1 esmolol, in random order. The lamb was allowed to stabilize at each inotropic state for at least 15 minutes (as determined by heart rate, LV pressure, and stroke volume), and a period of 20 to 30 minutes was allowed between conditions for return to baseline. Under each condition, the lamb was paced at various heart rates and modes to compare the effects of changes in frequency and mode of conduction on contractility between the groups (Table 1⇓).
The pacing modes included atrial or AV pacing and ventricular pacing at 130, 180, and 250 bpm. It was not possible to attain all states in all lambs. For example, during dobutamine infusion, the control lambs’ native heart rate often rose to 220 to 230 bpm, making atrial pacing at 180 bpm impossible. Furthermore, the resting heart rate of our control neonatal lambs at baseline was often in the range of 160 to 200 bpm; thus, lower heart rate interventions were not always possible. Finally, AV pacing could be performed only at a maximal rate of 180 bpm, with an AV delay set arbitrarily at 0.12 ms because of the limitations of available AV sequential pacemakers or pulse generators.
At the completion of the study, the lamb was euthanatized with high-dose pentobarbital injection. After euthanasia, the hearts were harvested, preserved in a 10% formalin solution, and submitted to pathology for evaluation of evidence of gross morphological abnormalities. The ventricles were divided into right ventricular and LV portions in a uniform fashion by a single operator. The interventricular septum was included with the left ventricle. The combined and individual ventricular weights were recorded. Tissue sections from the right and left ventricles and the interventricular septum were stained with hematoxylin and eosin for light microscopic analysis. The region of the AV node was dissected out and stained with trichrome and hematoxylin and eosin.
From the data acquired to disk, contractility was estimated by the end-systolic pressure-volume relationship by analyzing the pressure-volume loops during IVC occlusion. End systole was defined as the point of maximal elastance with a time-varying elastance model: [P(t)]/V(t)−Vd, where P is pressure, V is volume, and Vd is the “unstressed” volume of the ventricle (ie, the volume at which ventricular chamber pressure is zero at end systole). Vd cannot be measured directly and therefore was determined by an alternative technique.10 From the end-systolic pressure-volume points thus acquired, the slope of the relationship (Ees) was determined and used as the index of contractility. Vd was not used because linear extrapolation to zero pressure is not reliable and because we did not correct for parallel conductance. Measured volumes in our experiment excluded true volumes by the conductance of structures contiguous to the LV cavity (“parallel conductance”), but this “parallel element” is relatively constant in the small heart across the cardiac cycle12 and thus would not affect the slope of the pressure-volume relationship.
Statistical analyses were performed by use of SPSS 6.0a for Windows. Baseline hemodynamic data were compared between groups by use of unpaired Student’s t test. Statistical significance was assumed at P<.05. All data are presented as mean±SD unless otherwise noted.
The effects of inotropic and pacing interventions on the two groups of lambs were determined by multiple linear regression by use of dummy variables with effects coding: where Y is the dependent variable of interest (Ees or cardiac output ), ao is the intercept or overall mean value of that variable, G is the dummy variable for the animal group (assigned +1 if CAVB, −1 if control), ag is the coefficient of effect on Y, P is the pacing technique (+1 if ventricular and −1 if atrial or AV sequential), D is a set of two variables for inotropic state ([1,0] if dobutamine, [0,1] if esmolol, and [−1,−1] if control), R is the heart rate range ([1,0] if 165<heart rate<210, [0,1] if heart rate >210, and [−1,−1] if heart rate <165), L1 is the group of CAVB animals ([1,0,0,0,0,0,0] for the first through [0,0,0,0,0,0,1] for the sixth and [−1,−1,−1,−1,−1,−1,−1] for the seventh), L2 is the control group of 10 lambs similarly coded, and the subsequent variables are the potential interactions between variables.13 When cardiac output was solved for, all variables relating to heart rate were dropped from the equation because cardiac output is defined in part by heart rate, so the variables would covary with the dependent variable.
After regression analysis, an independent variable was considered “reliable” if the correlation coefficient was ≥.5 and the value for the t statistic of the slope of the line was P<.01. Statistical significance was assumed for dependent variables and sets of dependent variables at P<.05 for both t and partial f statistics. Unpaired t tests were used to compare lamb weights and postmortem heart weights.
Of the 34 animal preparations of CAVB, there were 9 fetal deaths. Two ewes were euthanatized because of postoperative infections. Three fetuses were delivered dead at term with a viable sibling. None had grown significantly from the time of surgery. Three aborted within 10 days of creation of CAVB. One ewe delivered twins 1 week before her due date; the CAVB fetus was dead without evidence of decompensation or decomposition, and the non-CAVB fetus survived. Of the 25 term deliveries, there were 4 perinatal deaths. All 4 were of full-term size (4.0 to 5.7 kg), and all were free of evidence of heart failure. Two of these were found to have intracranial bleeds, and 2 had no identifiable cause of death despite postmortem examination. Twelve other lambs that were delivered alive had converted out of CAVB before study. The 9 remaining fetuses were delivered at term and remained in CAVB.
All study lambs were assessed for general health status by the veterinary staff at the University of California, San Francisco. Compared with control lambs, all were of appropriate size and weight at birth. None were born with any signs or symptoms of hydrops fetalis. All had healed thoracotomy incisions and were free of infection. During the first week of life, all lambs gained weight and did not develop any signs or symptoms that suggested inadequate cardiac output. At ≈1 week of life, we studied 7 of the 9 chronically paced lambs (2 were not studied because of hardware malfunction) and 10 age-matched control lambs according to the protocol described above.
Baseline Hemodynamic Data
Hemodynamic data were obtained from 7 chronically paced and 10 age-matched control lambs at ≈1 week of life. Table 2⇓ summarizes the baseline data attained from both groups under their native conditions (ie, ventricularly paced in the study lambs and normal, sinus rhythm in the control lambs).
There was no significant difference between groups of the mean right and left atrial pressures or the peak LV pressures. Because we paced the CAVB lambs at a rate as close as possible to a true baseline state in utero, the mean heart rate in the chronically paced group of lambs is significantly lower than that of the normal lambs. This was associated with a lower cardiac output but a similar stroke volume between groups. Contractility was also not different between groups.
The overall mean cardiac output was 840 mL/min. The CAVB lambs were predicted to have an output that was 50 mL less than the overall mean or 12% less than the control lambs over all conditions. Ventricular pacing was associated with an output ≈12% less than an atrial rhythm overall, but this effect was less in the CAVB lambs (P=.008). There were significant and similar increases in output with β-adrenergic stimulation (21% greater than control) and decreases with β-adrenergic blockade (28% less than control). The effect of the dobutamine was greater in the ventricularly paced lambs of both groups compared with atrial rhythms (P=.04).
Overall, there were no significant differences in Ees between the control and CAVB lambs. Pacing site did not affect Ees. Ees did not significantly increase with dobutamine but did decrease with esmolol to 27% less than control. The highest heart rates were associated with a greater Ees. There were no interactions across groups, type of pacing, inotropic state, or heart rate (Table 3⇓).
Gross Morphology and Histology
There was no evidence of any gross morphological abnormalities on examination. There was no difference in the right ventricular or LV weights between the groups (Table 4⇓). Hematoxylin and eosin staining of the myocardium showed normal ventricular myocardium without evidence of myofibrillar disarray. Trichrome staining of the nodal area showed complete replacement of the conduction tissue by collagenous scar tissue.
With the advent of new echocardiographic and Doppler ultrasound techniques, CAVB is now diagnosed prenatally more often.2 3 4 5 CAVB occurs in 1 in 20 000 births14 in association with either fetal cardiomyopathy caused by anti-Ro antibodies or structural congenital heart disease. One consequence of CAVB in the fetus is the development of hydrops fetalis,15 which occurs in ≈30% of cases of antibody-associated CAVB and 50% of cases associated with structural heart defects.1 The presence of hydrops is a grave prognostic sign, with fetal demise approaching 100%.
The mechanism for the development of hydrops fetalis in both settings is inadequate cardiac output secondary to low heart rates, and in the setting of autoimmune cardiomyopathy, there may be an associated depression of myocardial function. Although the Frank-Starling mechanism is functional in the fetus, alterations in heart rate appear to have a greater effect on cardiac output than do adjustments of either preload or afterload.16 17 18 Coordinated atrial contraction also contributes more to ventricular filling in the fetus than after birth.19 20 In fetuses with CAVB, perinatal mortality correlates with atrial and ventricular rates.1 Even though the low cardiac output, and therefore hydrops, is due in part to the depressed myocardial function associated with the cardiomyopathy or the heart defect, the low heart rate contributes independently to the low output in both settings. Although pacing will not cure the cardiomyopathy or the structural defect, it may be therapeutic, allowing the fetus to survive these other abnormalities. Therefore, one aspect of the treatment of hydrops fetalis in the presence of CAVB should be aimed at augmenting cardiac output by increasing heart rate.
Currently, there are two accepted modalities to augment the heart rate in the hydropic fetuses with CAVB. The first is the administration of sympathomimetic drugs to the mother; these agents cross the placenta and increase fetal heart rate in some cases. This treatment exposes the mother to potent pharmacological agents and results in an unpredictable change in fetal heart rate. The second treatment is the emergent delivery of the fetus by cesarean section and placement of a cardiac pacemaker. This, however, is attended by the numerous complications associated with delivery of a premature infant in heart failure. The overall mortality with these two methods of treatment is ≈80%.21 22 An alternative approach to this problem is the surgical placement of a cardiac pacemaker on the surface of the fetal heart. In studies in our laboratory, this approach has recently been shown to be technically feasible in the fetal lamb model.23 This technique has the potential to provide a reliable and significant increase in heart rate, avoiding the aforementioned complications of the other treatment options.
In our study, the overall fetal loss rate (13/34=38%) is consistent with laboratory experiences with complex fetal models. It is difficult to fully understand what this fetal loss rate means. Certainly, a number of deaths within the first week or so after the AV block is created and pacing is initiated are expected, given the complexity of the surgery to create the model. From our data, however, it is impossible to say which part of the procedure led to their death. Similarly, it is debatable whether the late-term fetal deaths and perinatal deaths were due to failure of the paced ventricle to provide an adequate cardiac output. Postmortem examination of the abortus did not provide any insights into the cause of death because of in utero decomposition, and as mentioned, except for the two neonates with intracranial bleeds, no further evidence of the cause of death was identified on postmortem examination of the two other neonatal deaths. CAVB was created by cryoablating the AV node as previously described by Assad et al.6 We found, as did they, that this technique abolishes the ventricular escape rhythm, thus making it impossible to include a study group of unpaced fetuses. The fact that only 9 of the 21 newborn lambs remained in AV block after delivery reflects the inadequacy of our present external cryoablation protocol to consistently create long-term, irreversible CAVB. However, this does not negatively affect our ability to adequately address the proposed aims of the study in the lambs that remained in block.
The chronically paced lambs appeared to be healthy. All were of appropriate size and weight, were free of clinical signs and symptoms of hydrops, and had healed surgical incisions, implying at least adequate perfusion to thrive during the prenatal and perinatal periods.
The baseline hemodynamic data were similar between the two study groups except for heart rate and, consequently, cardiac output. The difference in the heart rate is inherent in the study protocol in that the baseline heart rate in the chronically paced lambs is limited by the maximum heart rate attainable from commercially available implantable pacemakers. There was no difference in the contractility or stroke volume; thus, the lower cardiac output was not a result of impaired ventricular function secondary to chronic pacing. Furthermore, there is no difference in contractility when hearts were paced synchronously compared with asynchronously.
To ensure that prenatal pacing did not adversely affect the ability of the myocardium to respond appropriately to inotropes (contractile reserve) and changes in heart rate, we administered a positive and a negative inotrope at various heart rates as described in the protocol. Contractility did not increase with dobutamine in either group of lambs, which is compatible with previous studies that showed an absence of contractile reserve in 1-week-old lambs.24 In both groups, esmolol decreased contractility to a similar extent. Furthermore, in both groups, the highest heart rates were associated with increased contractility, indicating an intact force-frequency relationship.
We submitted the heart specimens to pathology for gross and histological examination. Chronic VVI pacing has been noted to adversely affect the ventricular histology of the immature canine heart.25 Our specimens did not show any myofibrillar disarray or other histological evidence of adverse effects of VVI pacing on the myocardium. However, further research needs to address this important issue.
The risk of this proposed approach is primarily that of exposing the mother to laparotomy and hysterotomy. The chance of inducing labor is a real consideration26 ; however, this poses no more risk than the standard treatments, including premature delivery. Experience with fetal surgical intervention has indicated that maternal complications are few and rarely significant.26 However, the potential sequelae such as the possible inability for future vaginal deliveries and the known adverse effects of tocolytic therapy should be addressed as very real concerns. The risks to the fetus are the potential complications of the surgical procedure, which is relatively simple technically compared with other fetal surgical procedures.
Given the high mortality when the current treatment options are used for nonimmune hydrops fetalis, it is probable that direct fetal cardiac pacing will soon be clinically applicable. Although it has been shown both to be technically feasible and to result in a significant and predictable rise in fetal heart rate, many questions concerning the developmental consequences of intrauterine cardiac manipulation remain to be answered.
This study investigated the morphological and functional consequences of direct epicardial pacing on the rapidly developing, immature fetal heart with CAVB and found that cardiac output was indeed maintained at a reasonable range and that there were no apparent significant adverse morphological consequences. However, because many anomalies are thought to arise from abnormal intracardiac blood flow, the lack of AV synchrony and the abnormal ventricular conduction pattern caused by epicardial ventricular electrode pacing may profoundly alter intracardiac and regional blood flow, which may in turn affect the morphological development of the fetal heart and other organ systems. Furthermore, in our model, the pacemaker lead was directly applied to the epicardium through a thoracotomy to ensure reliable lead placement and improve ventricular capture. However, there are obvious disadvantages to such an extensive approach in the young fetus, and other approaches such as transvenous lead placement should be investigated. Additionally, although increasing the heart rate in the fetus with CAVB seems compellingly logical, no absolute proof exists that simply increasing the ventricular rate will provide adequately beneficial augmentation of cardiac output in the presence of autoantibody-induced or structurally related cardiomyopathy. Therefore, further investigation of these issues needs to be undertaken before application of this approach in the human fetus.
In summary, results of treating prenatally diagnosed congenital CAVB with current therapeutic options are dismal. Therefore, we have investigated the efficacy of direct epicardial pacing in a fetal lamb model. This model is somewhat limited in that it is a created state of CAVB in a previously healthy fetus at 80% gestation and that epicardial pacing is begun before the fetus develops distress or hydrops. However, the results indicate that in this model, the technique of prenatal epicardial pacing is well tolerated by the fetus, can be successfully applied for the treatment of CAVB, and does not adversely affect myocardial function in the rapidly developing, immature heart.
This work was supported in part by NIH grants HL-43357 and HL-08824-02.
Reprint requests to V. Mohan Reddy, MD, 505 Parnassus Ave, M593, San Francisco, CA 94143-0118.
- Received April 8, 1996.
- Revision received January 22, 1997.
- Accepted January 23, 1997.
- Copyright © 1997 by American Heart Association
Schmidt KG, Ulmer HE, Silverman NH, Kleinman CS, Copel JA. Perinatal outcome of fetal complete atrioventricular block: a multicenter experience. J Am Coll Cardiol. 1991;17:1360-1366.
Kleinman CS, Donnerstein RL, Jaffe CC. Fetal echocardiography: a tool for evaluation of in utero cardiac arrhythmias and monitoring of in utero therapy: analysis of 71 patients. Am J Cardiol. 1983;51:237-243.
Allan LD, Anderson RH, Sullivan ID, Campbell S, Holt DW, Tynan M. Evaluation of fetal arrhythmias by echocardiography. Br Heart J. 1983;50:240-245.
Silverman NH, Enderlein MA, Stanger P, Teitel DF, Heymann MA, Golbus MS. Recognition of fetal arrhythmias by echocardiography. J Clin Ultrasound. 1985;13:255-263.
Strasburger JF, Huhta JC, Carpenter RJ, Garson A Jr, McNamara DG. Doppler echocardiography in the diagnosis and management of persistent fetal arrhythmias. J Am Coll Cardiol. 1986;7:1386-1391.
Assad RS, Aillo VD, Jatene MB, Costa R, Hanley FL, Jatene AD. Cryosurgical ablation of fetal atrioventricular node: new model to treat fetal malignant tachyarrythmias. Ann Thorac Surg. 1995;60:S629-S632.
Baan J, van der Velde ET, De Bruin HG, Smeenk GJ, Koops J, van Dijk AD, Temmerman D, Senden J. Buis B. Continuous measurement of left ventricular volume in animals and humans by conductance catheter. Circulation. 1984;70:812-823.
Baan J, van der Velde ET, Steendijk P, Koops J. Calibration and application of the conductance catheter for ventricular volume measurement. Automedica. 1989;11:357-365.
Kass DA, Maughan WL, Guo ZM, Kono A, Sunagawa K, Sagawa K. Comparative influence of load versus inotropic states on indexes of ventricular contractility: experimental and theoretical analysis based on pressure-volume relationships. Circulation. 1987;76:1422-1436.
Applegate RJ, Cheng CP, Little WC. Simultaneous conductance catheter and dimension assessment of left ventricular volume in the intact animal. Circulation. 1990;81:638-648.
Teitel DF, Klautz RK, Steendijk P, van der Velde ET, van Bel F, Baan J. The end-systolic pressure-volume relationship in the newborn lamb: effects of loading and inotropic interventions. Pediatr Res. 1991;29:473-482.
Cassidy SC, Teitel DF. The conductance volume catheter technique for the measurement of left ventricular volume in young piglets. Pediatr Res. 1992;31:85-90.
Glantz SA, Slinker BK. Primer of Applied Regression and Analysis of Variance. New York, NY: McGraw-Hill Publishing Co; 1990:329-349.
Michaelsson M, Engle MA. Congenital complete heart block: an international study of the natural history. Cardiovasc Clin. 1972;4:85-101.
Altenburger KM, Jedziniak M, Roper WL, Hernandez J. Congenital complete heart block associated with hydrops fetalis. J Pediatr. 1977;91:618-620.
Rudolph AM, Heymann MA. Cardiac output in the fetal lamb: the effects of spontaneous and induced changes of heart rate on right and left ventricular output. Am J Obstet Gynecol. 1976;124:183-192.
Thornburg KL, Morton MJ. Filling and arterial pressures as determinants of right ventricular stroke volume in the sheep fetus. Am J Physiol. 1983;244:H656-H663.
Hawkins JA, Van Hare GF, Schmidt KG, Rudolph AM. The effects of increasing afterload on left ventricular output in fetal lambs. Circ Res. 1989;65:127-134.
Reed KL, Sahn DJ, Scagnelli S, Anderson CF, Shenker L. Doppler echocardiographic studies of diastolic function in the human fetal heart: changes during gestation. J Am Coll Cardiol. 1986;8:391-395.
Crombleholme TM, Longaker MT, Langer JC, Bradley SM, Duncan BW, Adzick S, Verrier ED, Harrison MR. Complete heart block and AV-sequential pacing in fetal lambs: the atrial contribution to combined ventricular output in the fetus. Surg Forum. 1989;40:268-271.
Kleinman CS, Donnerstein RL. Ultrasound assessment of cardiac function in the intact human fetus. J Am Coll Cardiol. 1985;5:84S-94S.
Holgreve W, Curry CJR, Golbus MS, Callen PW, Filly RA, Smith JC. Investigation of nonimmune hydrops fetalis. Am J Obstet Gynecol. 1984;150:805-812.
Heinemann MK, Sabik JF, Gamble WJ, Hanley FL. Prenatal pacemaker implantation in the fetal lamb. Presented at the 64th Scientific Sessions of the American Heart Association: November 13, 1991; Anaheim, Calif.
Teitel DF, Sidi D, Chin T, Brett C, Heymann MA, Rudolph AM. Developmental changes in myocardial contractile reserve in the lamb. Pediatr Res. 1985;19:948-955.
Karpawich PP, Justice CD, Cavitt DL, Chang C. Developmental sequelae of fixed-rate ventricular pacing in the immature canine heart: an electrophysiologic, hemodynamic, and histopathologic evaluation. Am Heart J. 1990;119:1077-1083.
Harrison MR, Adzick NS, Longaker MT, Goldberg JD, Rosen MA, Filly RA, Evans MI, Golbus MS. Successful repair in utero of a fetal diaphragmatic hernia after removal of herniated viscera from the left thorax. N Engl J Med. 1990;322:1582-1584.