WITH childhood cancer survival rates on the rise, increasing numbers of female pediatric cancer patients are reaching childbearing age. The long-range effects of childhood chemotherapy can have significant clinical implications for these women. The toxicity of an antineoplastic drug given to the pediatric patient may not be fully apparent until other physical conditions in adulthood, such as pregnancy, force the issue. Pregnancy imposes significant physiologic changes and cardiovascular stress. The cardiotoxicity of one chemotherapeutic agent in particular, doxorubicin, is frequent in young females, potentially severely limiting myocardial growth in childhood, leading to a lifelong reduction in myocardial mass, which may result in a significantly decreased cardiac reserve in the child as she grows. 1,2Pregnancy can unmask the cardiovascular damages caused by doxorubicin even from the remote past in otherwise asymptomatic patients. When these women require an anesthetic for vaginal or cesarean delivery, anesthesiologists should be aware of the possibility of cardiac failure, even in asymptomatic pregnant patients, if they have previously been treated with doxorubicin.
The interaction of the long-term cardiotoxic effect of doxorubicin with pregnancy and anesthetic management has not been well described. We present two cases in which regional anesthesia, particularly combined spinal-epidural (CSE) technique, was used successfully for cesarean delivery and vaginal delivery in kyphoscoliotic patients with doxorubicin-induced cardiomyopathy. To our knowledge, this is the first report of the successful use of epidural and combined spinal-epidural technique in kyphoscoliotic pregnant patients with doxorubicin-induced cardiomyopathy presenting during pregnancy.
A 35-yr-old kyphoscoliotic woman, primigravida, at 36 weeks of gestation with fetal breech presentation was admitted to the intensive care unit (ICU) for pulmonary edema and congestive heart failure. Her medical history was significant for osteosarcoma, which was treated with left hemipelvectomy and intravenous doxorubicin 20 yr earlier. She was active and asymptomatic with good exercise tolerance until 28 weeks of gestation, when she developed shortness of breath and dyspnea. Echocardiography revealed moderate mitral regurgitation and a 30% ejection fraction, which was improved with digoxin 0.25 mg/day. By 36 weeks of gestation, her clinical status deteriorated and led to this ICU admission. Repeated echocardiography revealed an ejection fraction of 10%, mild mitral and moderate tricuspid regurgitation, dilated ventricles with systolic dysfunction, and mild pulmonary hypertension. She improved clinically with diuresis, but required maximum doses of milrinone (0.75 μg·kg−1·min−1) and dobutamine (15 μg·kg−1·min−1) to maintain her hemodynamic stability. She later required the switching of dobutamine to dopamine infusion (10– 15 μg·kg−1·min−1) and the addition of lidocaine infusion (1 mg/min) to control ventricular dysrythmias. To prevent cardiovascular deterioration, a semiurgent cesarean section was planned. In the ICU, her pulmonary artery catheter revealed CO of 3.5–4.2 l/min, systemic vascular resistance of 950–1550 dynes.s.cm−5, central venous pressure of 13–16 mmHg, PAS of 40 to 50 mmHg, PAD of 20 to 36 mmHg, and pulmonary capillary wedge pressure of 16–18 mmHg. Her blood pressure was 115/60 with a sinus tachycardia rate of 130, and her oxygen saturation was 96–98% while on 6 l/min of oxygen by nasal cannula. A low dose combined spinal epidural technique was performed (intrathecal 0.8 ml of hyperbaric 0.75% bupivacaine, 25 μg fentanyl, and 50 μg epinephrine, then followed by titrating epidurally 9 ml of 2% lidocaine with epinephrine). With a bilateral sensory anesthesia level of T5, she was maintained in a semiupright position of 30° throughout the epidural dosing and the surgical procedure to allow better spontaneous ventilation and to limit the upper cephalad spread of the intrathecal local anesthetic. She received 150 ml of 25% albumin during and after the placement of the combined spinal-epidural anesthetic. In addition to inotropes, she also required small amounts of phenylephrine (20 μg increments, total of 120 μg) after the delivery of the fetus to correct hypotension. The patient and the fetus tolerated the anesthetic and surgical procedure with no complications and with an estimated blood loss of 1,400 ml. The patient was delivered of a live infant of 2,840 g with 1 and 5 min APGAR scores of 9. She received epidural patient-controlled analgesia with fentanyl for postoperative pain management. She was weaned off inotropes in 24–48 h and discharged without complications on the third postpartum day with oral digoxin and captopril. At her 6-week postpartum visit, she remained asymptomatic and had no other complications.
A 31-yr-old kyphoscoliotic woman, G2P1, at 35 weeks of gestation was admitted to our hospital with progressive shortness of breath. Her medical history was significant for Ewing's sarcoma at age 16, which was treated with intravenous doxorubicin and resection of her right clavicle. She had been active and asymptomatic until the fifth postpartum day after her first vaginal delivery at age 24, at which time she developed pulmonary edema with an ejection fraction of 35%. She recovered uneventfully with ejection fraction improved to 40–45% after receiving diuresis and vasodilator. She remained asymptomatic for 4 yr until the thirty-fifth week of gestation of her second pregnancy when she developed pulmonary edema. Echocardiography showed a decreased ejection fraction of 35%, mild pulmonary hypertension, mild tricuspid and mitral regurgitation, biatrial enlargement, a dilated right ventricle, and a hypokinetic left ventricle. She responded well and improved upon diuresis. An epidural catheter was placed with moderate technical difficulty during her labor induction. A low concentration epidural local anesthetic (0.125% bupivacaine) with fentanyl (1.5 μg/ml) was administered successfully for early labor analgesia. A significantly higher volume and concentration of a local epidural anesthetic and narcotic was required for the later stage of labor. A healthy infant with APGAR scores of 9 was delivered uneventfully. The patient remained asymptomatic until her third pregnancy 4 yr later.
The above patient, at age 35, G3P2, (19 yr from doxorubicin chemotherapy) at 35 weeks of gestation was admitted to our hospital with the diagnosis of pulmonary edema and ejection fraction of 25%. She responded well and improved upon diuresis, and induction of labor was initiated. Based on her positive fluid balance, no preload was administered before anesthetic. A combined spinal-epidural (CSE) technique was used to administer intrathecal sufentanil 7.5 μg, 1.25 mg bupivacaine, and 50 μg epinephrine. She did not develop hypotension. A healthy infant was delivered uneventfully with normal APGAR scores. At her 6-week postpartum visit, she remained asymptomatic without complications.
The anthracycline chemotherapeutic agents, doxorubicin (Adriamycin) and daunorubicin (Cerubidine) intercalate with DNA, affecting DNA and RNA synthesis. The anthracyclines react with cytochrome P450 reductase in the presence of reduced nicotinamide adenine dinucleotide phosphate (NADPH) to form semiquinone radical intermediates, which in turn react with oxygen to produce superoxide anion radicals (OH). These radicals are highly destructive to cells, including myocytes. Several clinical presentations of doxorubicin-induced cardiotoxicity, including atrial and ventricular dysrhythmias, a pericarditis-myocarditis syndrome, acute hypertensive reactions, cardiovascular collapse, and death during and after administration of anesthetics, have been described in nonpregnant patients. 3,4,5
Pregnancy is accompanied by a significant blood-volume expansion, an increase in metabolic demands, dilution anemia, and alterations in vascular resistance that are associated with ventricular dilatation and increases in cardiac output. 6A patient treated with doxorubicin in childhood may have a significantly reduced myocardial mass and decreased cardiac reserve, but may remain asymptomatic, developing heart failure for the first time during pregnancy, especially during the third trimester and the immediate postpartum period when the cardiac output requirement and demand are highest. Both patients reported here had been asymptomatic and active with good exercise tolerance for many years after doxorubicin therapy. Only in their third trimesters of pregnancy did they experience any cardiac symptoms.
In pregnant patients with cardiomyopathy alone, the optimum anesthetic technique for cesarean section is controversial, and both general and regional anesthesia techniques have been described. 7–10Even more challenging, however, is performing a rapid sequence induction on a pregnant kyphoscoliotic patient with severely compromised cardiac function. It can be potentially detrimental to both mother and fetus. Several authors have reported perioperative cardiovascular collapse, and complications, and even death in nonpregnant patients who previously received doxorubicin who are administered general anesthesia. 3–5McIncoe reported a case of peripartum cardiomyopathy presenting as a cardiac arrest at induction of general anesthesia for cesarean section. 11A retrospective analysis carried out in the United States from 1979–1990, showed that most maternal deaths resulting from anesthesia-related complications occurred during general anesthesia for cesarean section. 12Furthermore, the risk ratio for general anesthesia had increased to 16.7 times that for regional anesthesia. A controlled afterload reduction by regional anesthesia may be beneficial for patients with low cardiac output.
Our first patient presented an unusual combination of anesthetic considerations: severe doxorubicin-induced cardiomyopathy, thoracolumbar kyphoscoliosis, and previous hemipelvectomy, in addition to the physiologic changes of pregnancy. We chose combined spinal-epidural (CSE) technique for the cesarean delivery because (1) CSE avoids the risk of general anesthesia, particularly its cardiac depressive effects in this patient with doxorubicin-induced cardiomyopathy; (2) CSE allows the use of a lower initial dose of intrathecal local anesthetic and a gradual setup of the final desired anesthetic level by titrating a small amount of epidural drug, thereby minimizing hypotension or hemodynamic changes and lowering the maternal and umbilical cord blood concentration of local anesthetics, 13all of which are critical for a patient with severe cardiomyopathy; (3) CSE allows flexibility for the duration of anesthesia, especially in potentially difficult surgical cases, such as this patient with a previous hemipelvectomy; (4) CSE allows postoperative use of epidural patient-controlled narcotics for postoperative pain management to minimize hemodynamic changes, postoperative stress, and respiratory depression of systemic narcotics; (5) CSE provides a denser quality of sensory block than epidural anesthetic alone and it has a lower failure rate versus epidural alone, 13,14,15,16especially in kyphoscoliotic patients, by the clear identification and confirmation of the intrathecal space and possibly the epidural space as well.
For this patient, we chose a low initial dose (6 mg) of intrathecal bupivacaine administered in the sitting position to minimize rapid cephalad spread and hypotension, and to provide dense anesthesia for the lower dermatomal level to approximately T10 sensory level, where the predominate part of the surgery took place. No vasopressor was needed during the establishment of the anesthetic block, and only a small amount of vasopressor was needed at the latter part of the surgery after the delivery of the fetus.
When the second patient presented for her third delivery, we also performed a CSE technique for many of the same reasons discussed above, especially because of the potential for diminished efficacy of the epidural during the later stage of her first labor anesthetic. Continuous spinal technique is also a possible option for these patients. 17The major disadvantage, particularly in the obstetric population, is the high potential for postdural puncture headache and difficulty in performing an epidural blood patch if subsequently required.
It is important for anesthesiologists to be aware of the possibility of cardiac failure in asymptomatic pregnant patients previously treated with doxorubicin even in the remote past. We believe that the administration of epidural and combined spinal-epidural anesthesia to patients with doxorubicin-induced cardiomyopathy for cesarean or vaginal delivery can be done safely with excellent outcomes and stable hemodynamics if it is performed with precision and meticulous attention to details in intravascular volume, anesthetic block level, drug dose, and hemodynamics.