Safety of intra-amniotic digoxin administration before late second-trimester abortion by dilation and evacuation Eleanor A. Drey, MD,a Lisa J. Thomas, MD,a Neal L. Benowitz, MD,b Nora Goldschlager, MD,c and Philip D. Darney, MD, MSca San Francisco, California OBJECTIVE: The purpose of this study was to determine the safety of intra-amniotic digoxin injection before late second-trimester pregnancy termination by dilation and evacuation through an assessment of maternal systemic digoxin absorption, cardiac rhythm, and coagulation parameters. STUDY DESIGN: Pregnant women at between 19 and 23 weeks’ gestation received 1.0 mg digoxin through intra-amniotic injection and then had serum digoxin levels determined for 48 hours and Holter cardiac monitoring performed for 24 hours. Clotting parameters were assessed before digoxin injection and 24 hours later, at the time of the dilation and evacuation procedure. RESULTS: Eight patients completed the study. The mean (±SD) serum digoxin peak concentration was 0.81 ± 0.22 µg/L (range, 0.5-1.1 µg/L). The mean (±SD) time to peak digoxin concentration was 11.0 ± 5.55 hours (range, 4-20 hours). Ambulatory cardiac monitoring showed no rhythm or conduction abnormalities associated with digoxin. Prothrombin time, partial thromboplastin time, and fibrinogen levels did not change significantly between determinations before and after the dilation and evacuation procedure (11.5 to 11.4 seconds, 24.1 to 24.4 seconds, and 441 to 475 mg/dL, respectively). CONCLUSION: The maximum digoxin concentration peak achieved after intra-amniotic injection was in the low therapeutic range. No rhythm or conduction abnormalities associated with digoxin were noted by Holter monitoring. Coagulation parameters did not change significantly. On the basis of the limited systemic absorption and the absence of clinically significant cardiac or clotting effects, intra-amniotically administered digoxin may be considered safe for use before late second-trimester pregnancy terminations. (Am J Obstet Gynecol 2000;182:1063-6.)
Key words: Digoxin, dilation and evacuation, second-trimester abortion, therapeutic abortion
More than 140,000 second-trimester abortions are performed annually in the United States, 94% of which are by dilation and evacuation.1 Second-trimester abortions account for a disproportionate burden of the morbidity and mortality related to abortion, with the risks of complications rising with each subsequent week of pregnancy. For example, the mortality risk associated with abortions performed at ≤8 weeks’ gestation is 0.4 per 100,000 procedures, compared with a risk of 10.4 per 100,000 procedures associated with abortions performed at ≥21 weeks’ gestation.2 Clinicians have been using digoxin to facilitate secondtrimester pregnancy termination for several years. Digoxin injection causes fetal death and is believed to
From the Center for Reproductive Health Research and Policy, Department of Obstetrics, Gynecology, and Reproductive Services, University of California, San Francisco,a and the Division of Clinical Pharmacology, Departments of Medicine and Biopharmaceutical Sciences,b and the Division of Cardiology, Department of Medicine,c San Francisco General Hospital. Supported by the Department of Obstetrics, Gynecology, and Reproductive Sciences, San Francisco General Hospital. Reprints not available from the authors. Copyright © 2000 by Mosby, Inc. 0002-9378/2000 $12.00 + 0 6/1/105438 doi:10.1067/mob.2000.105438
make the dilation and evacuation procedure easier and safer because the fetal tissue is softened. Another advantage is that both the patient and the clinician may prefer to abort a dead fetus. Digoxin has been administered by the intracardiac, intrathoracic, intrafetal, and intra-amniotic routes, with doses varying from 0.25 to 2 mg.3 Most clinicians who use digoxin usually inject it 1 to 2 days before the dilation and evacuation procedure, at the time of laminaria placement. Although there are no reports of maternal side effects or complications as a result of this use of digoxin, neither are there any evaluations of digoxin’s safety or efficacy before dilation and evacuation. The purpose of this study was to assess the safety of intra-amniotic administration of digoxin before late second-trimester dilation and evacuation by evaluating its systemic absorption and its effects on cardiac rhythm and conduction and on coagulation measurements. Methods After approval by the committee on human research, a safety study was conducted at San Francisco General Hospital between July and September 1998. Women were eligible who had elected to undergo pregnancy termination between 18 weeks and 23 weeks 1 day of gesta1063
1064 Drey et al
May 2000 Am J Obstet Gynecol
Fig 1. Changes in serum digoxin concentration through 48 hours after intra-amniotic administration of 1 mg digoxin (n = 8 women). Data points, Mean; error bars, SD.
Table I. Frequencies and types of arrhythmias and heart block occurring in 24 hours after intra-amniotic digoxin administration in preparation for dilation and evacuation (n = 8 women) Rare Occasional Frequent (0-30 episodes/h) (31-60 episodes/h) (>60 episodes/h) Premature atrial complexes Premature ventricular complexes Junctional tachycardia* Atrial tachycardia† Atrioventricular block
7
1
0
7
0
0
1 1 1
0 0 0
0 0 0
All figures represent numbers of patients. *There was a single 2-second episode of junctional tachycardia at 125 beats/min. †There was a single 1-second episode of atrial tachycardia at 140 beats/min.
tion, as dated by ultrasonographic biparietal diameter measurement. Eight patients were selected as an adequate sample to characterize the systemic exposure to digoxin with a reasonable degree of confidence. To be included in the study women had to be ≥18 years old, had to be able to give informed consent, and had to have documented fetal cardiac activity. Women were excluded because of significant medical illness or cardiovascular disease, current use of cardiac or antihypertensive medications, a known digoxin allergy, pregnancy complications (oligohydramnios, polyhydramnios, multiple gestation, or congenital anomalies), maternal weight >30% above ideal, difficult maternal venous access, or abnormal serum potassium levels (<3.5 mmol/L or >5 mmol/L).
After providing consent, participants underwent a physical examination, Holter monitor placement, and blood sampling for type, complete blood cell count, and determination of prothrombin time, partial thromboplastin time, fibrinogen concentration, potassium concentration, and digoxin level. Two potential subjects were excluded because of hypokalemia. The 8 remaining subjects received 1.0 mg digoxin intra-amniotically through amniocentesis with a 22-gauge spinal needle, which was performed under ultrasonographic guidance to avoid transplacental placement of the needle. After a paracervical block with 10 mL 1% chloroprocaine, several laminaria were placed in the internal os. After digoxin administration, subjects were monitored 24 hours for toxicity. Their vital signs were checked and digoxin levels were assessed every 4 hours. The Holter monitor was placed ≥1 hour before the digoxin injection and removed 24 hours after the initial amniocentesis. One of two providers (Eleanor A. Drey, MD, or Lisa J. Thomas, MD) performed the dilatation and evacuation procedure under ultrasonographic guidance. A sample of the amniotic fluid evacuated at the start of the dilation and evacuation procedure was evaluated for its digoxin content. One hour after the dilation and evacuation procedure the prothrombin time, partial thromboplastin time, fibrinogen concentration, and digoxin level were determined and subjects completed a questionnaire that included questions about digoxin-related side effects, including nausea, loss of appetite, fatigue, headache, confusion, blurred vision, lightheadedness, and seeing halos or light around objects. The complete blood cell count and digoxin level were checked again 48 hours after the amniocentesis, at which time patients received $100 for participation. Holter monitor recordings were analyzed
Drey et al 1065
Volume 182, Number 5 Am J Obstet Gynecol
Table II. Changes in coagulation measurements 24 hours after intra-amniotic digoxin injection in preparation for dilation and evacuation (n = 8 patients) Baseline
Prothrombin time (s) Partial thromboplastin time (s) Fibrinogen (mg/dL)
24 h after digoxin*
Mean
Range
Mean
Range
Difference
Statistical significance
11.5 24.1 441
10.8-12.3 21.3-27.6 303-583
11.4 24.4 475
10.9-11.9 21.3-29.8 365-628
0.1 0.3 34
NS NS NS
*All 24-hour values are based on 7 patients, because one sample was of insufficient quantity. NS, Not statistically significant.
by a trained cardiology technician. Results were reviewed by an attending cardiologist (Nora Goldschlager, MD), who reviews all Holter monitoring performed at San Francisco General Hospital and who was blinded to patients’ digoxin levels. The digoxin assay was performed with an Abbott TDFLx (Abbott Laboratories, Abbott Park, Ill) assay kit with a heterogeneous competitive magnetic separation assay. The protocol and consent form were approved by the University of California, San Francisco, Committee on Human Research. Results Eight patients were recruited, 3 of whom were white, 3 of whom were black, and 2 of whom were Native American. The mean age was 24.1 years (range, 18-35 years), mean gestational age was 21 weeks 2 days’ gestation (range, 19 weeks 6 days’–22 weeks 4 days’ gestation), mean weight was 151 lb (range, 98-222 lb), and mean body mass index was 25 kg/m2 (range, 19-34 kg/m2). Gravidity ranged from 2 to 8, parity ranged from 0 to 5, and previous elective abortions ranged from 0 to 1. The mean (±SD) peak digoxin level was 0.81 ± 0.22 µg/L (range, 0.5-1.1 µg/L; median, 0.8 µg/L). The mean (±SD) time from injection to peak digoxin concentration was 11.0 ± 5.55 hours (range, 4-20 hours; median, 12.0 hours; Fig 1). There was a trend toward decreasing digoxin levels with increasing body mass index, but this trend did not show statistical significance (r = –0.62; P = .1). Digoxin levels in amniotic fluid collected at the time of the dilation and evacuation procedure ranged widely, with a mean of 1150 µg/L and a range of 480 to 3780 µg/L. These measurements were calculated with serum as a reference because there are no standards for amniotic fluid digoxin levels. Ambulatory electrocardiographic monitoring (Table I) revealed 1 patient with occasional (31-60 episodes per hour) premature atrial complexes and rare premature ventricular complexes, with three episodes of vagally mediated first- and second-degree atrioventricular block, during which there was abrupt sinus rate slowing with nonconducted P waves. 4 One patient reported an episode of dizziness, during which she was in normal
Table III. Mean peak digoxin levels according to reported side effects after intra-amniotic administration of digoxin in preparation for dilatation and evacuation (n = 8)
Side effect Nausea Loss of appetite Headache Fatigue Confusion Blurred vision Lightheadedness Seeing halos or light around objects
No.
Mean peak digoxin level in patients with side effect (µg/L)
Mean peak digoxin level in patients without side effect (µg/L)
2 4 4 5 1 1 1 1
0.55 0.70 0.78 0.72 0.90 0.90 0.60 0.50
0.93 0.98 0.90 1.03 0.83 0.83 0.87 0.89
sinus rhythm. No other patients reported cardiac symptoms. The prothrombin time, partial thromboplastin time, and fibrinogen concentration all remained stable for 24 hours with respect to coagulation parameters before the digoxin injection (Table II). Reports of side effects associated with digoxin toxicity did not correlate with peak digoxin levels of the affected patients (Table III). Comment Although digoxin could have been administered in a variety of doses into either the amniotic fluid or the fetus, we administered a fixed dose of 1 mg intra-amniotically, because success with this route is easily confirmed by withdrawal of amniotic fluid. Eight subjects were enrolled, a number considered sufficient to demonstrate safety in a phase I clinical trial. The peak digoxin level occurred approximately 11 hours after drug administration (Fig 1). As one would predict in light of the small sample size, the SD was large. Even at its peak, however, digoxin concentration achieved a level only in the low therapeutic range, with the highest level seen, 1.1 µg/L, well below the level of 2 µg/L at which a patient might manifest toxicity. (A review of digoxin toxicity—usually defined by arrhythmias
1066 Drey et al
that disappeared when digoxin was withheld—mostly showed mean toxic levels to be ≥3.0 µg/L,5 levels 3 times greater than our observed peak.) The trend toward decreasing peak digoxin levels with increasing body mass index was not statistically significant. Ambulatory cardiac monitoring did not show any patterns that suggested digoxin toxicity. The rare premature atrial complexes, rare premature ventricular complexes, and single episode each of transient junctional and atrial tachycardias are of no clinical significance and are consistent with the rates and rhythms expected in a population of young women. The one noteworthy pattern, that of the patient with vagally mediated first- and second-degree atrioventricular block, has no established relation to digoxin exposure and is common among healthy individuals. The period of fetal death before the pregnancy termination (<24 hours) was not associated with significant changes in coagulation measurements (Table II). Because of the low serum digoxin levels, it is not surprising that the side effects that might suggest digoxin intoxication, most of which are relatively nonspecific, bore no relation to subjects’ digoxin levels (Table III). In support of the conclusion that symptoms were not related to digoxin effects, those patients who reported symptoms actually had lower average peak digoxin levels than did
May 2000 Am J Obstet Gynecol
patients without the same symptoms. Because of the small sample size, however, any conclusions about side effects may be subject to a type II error. We conclude that intra-amniotic administration of 1 mg digoxin before termination of pregnancy during the late second trimester does not result in clinically significant elevation of maternal serum digoxin levels, is not associated with evidence of digoxin toxicity, does not alter maternal cardiac rate or rhythm, and does not change clotting parameters. Although digoxin injection appears safe, determination of its clinical efficacy requires a randomized trial. REFERENCES
1. Abortion surveillance: preliminary data—United States, 1995. MMWR Morb Mortal Wkly Rep 1998;47(SS-2):31-40. 2. Lawson HW, Frye A, Atrash HK, Smith JC, Shulman HB, Ramick M. Abortion mortality, United States, 1972-1987. Am J Obstet Gynecol 1994;171:1365-72. 3. Hern WM, Zen C, Ferguson KA, Hart V, Haseman MV. Outpatient abortion for fetal anomaly and fetal death from 1534 menstrual weeks’ gestation: techniques and clinical management. Obstet Gynecol 1993;81:301-6. 4. Massie B, Scheinman MM, Peters R, Desai J, Hirschfeld D, O’Young J. Clinical and electrophysiologic findings in patients with paroxysmal slowing of the sinus rate and apparent Mobitz type II atrioventricular block. Circulation 1978;58:305-14. 5. Lee TH, Smith TW. Serum digoxin concentration and diagnosis of digitalis toxicity: current concepts. Clin Pharmacokinet 1983;8:279-85.