A 12-year-old girl first presented to her family physician in January 1991 with an episode of tachycardia and dyspnea. Her mother described the patient’s appearance during the event as “like a statue,” extremely pale, gray, and lethargic. An ECG at that time showed sinus tachycardia with a rate of 143 bpm and T-wave inversion in lead III (Fig 1⇓). Oxygen was administered in the office, and a trial of propranolol 10 mg/d was begun.
The girl’s symptoms had first occurred 6 months earlier and were of variable duration and severity. Tachycardia would persist for several hours and be associated with dyspnea and pallor. During the episode, she would develop generalized myalgia, weakness, and fatigue and then sleep for 2 to 7 hours. There was no associated chest pain, dizziness, or presyncope. Episodes occurred one to three times per week, in the afternoon, and were unrelated to activity. She had decreased exercise tolerance compared with her peers.
She was referred to a pediatric cardiologist for further evaluation. A Holter recording was normal. Her cardiac examination was normal; however, an echocardiogram reportedly showed discontinuity of the atrial septum, with left-to-right flow across the interatrial septum by pulsed Doppler and color flow Doppler. The diagnosis of secundum atrial septal defect was made and a cardiac catheterization recommended, but the family declined. Propranolol was discontinued, in part because of a history of asthma. Verapamil was started but was soon discontinued because of dizziness. The family then sought a second opinion.
The patient was the product of a 31-week gestation, with delivery by emergency cesarean section for preterm labor with fetal distress. Birth weight was 4 lb 1 oz. She remained hospitalized for 1 month as a neonate, requiring ventilation and oxygen administration. Development of motor skills was delayed; she walked at age 2.5 years and received physical therapy until the age of 10 years. At the time of the present illness, her motor skills were within the normal range. She had experienced frequent upper respiratory infections and had two isolated episodes of asthma, the major symptom of which was coughing. She had received immunoglobulin injections from age 3 to 4 years and allergy desensitization therapy from age 4 to 5 years.
Her mother had both Sjögren’s syndrome and symptomatic Wolff-Parkinson-White syndrome, well controlled with propranolol and digoxin. Her father had coronary artery disease for which he had undergone coronary artery bypass surgery at age 50. There was a maternal uncle with an unspecified tachycardia. She had no siblings.
She lived at home with her parents in a rural area surrounded by rice fields that are fertilized twice a year. The home was 3 miles from an industrial waste-treatment complex that processes solid waste material from the Houston Ship Channel through a series of connecting lagoons that lead to a treatment facility. The family and their neighbors were very concerned about the characteristic odor in the neighborhood associated with emissions from the treatment plant and about the noticeable lack of wildlife on the land, including rodents and roaches. Environmental activist groups and public health officials had investigated the facility, and litigation was pending. The family drank water from a well 120 ft deep; the well had been examined to exclude the possibility of contamination from groundwater sources. She attended school only half days because of fatigue. She consumed one or two caffeine-containing beverages per day but denied use of tobacco and recreational drugs. The only medication taken was acetaminophen, as needed. Her allergies were environmental.
On physical examination, the patient was well nourished and in no distress. She appeared sleepy and spoke slowly and deliberately. Her pulse was 96 bpm; blood pressure was 113/60 mm Hg in the right arm. Respiratory rate was 18 breaths per minute. Temperature was 98.4°F. Her height was 163 cm (95th percentile) and weight, 57 kg (90th percentile). There was no thyromegaly. Lungs were clear to auscultation. The precordium was quiet, and the apex beat was not displaced. S1 was normal. S2 split normally with respiration and was not increased in intensity. A grade 1/6 vibratory musical murmur was heard along the lower left sternal border, with no radiation. There were no diastolic murmurs. Pulses were full and symmetrical. The abdomen was soft, with no hepatosplenomegaly or bruits. Extremities were warm and well perfused, with no cyanosis, edema, clubbing, or pallor. Neurological examination was normal.
Initial laboratory evaluation showed a hemoglobin of 13.5 g%, with a reticulocyte count of 1.3% and 8300 white blood cells. Erythrocyte sedimentation rate was 20 mm/h. Thyroid function tests were normal. Epstein-Barr virus, antinuclear antibody, and rheumatoid factor screens were negative. Serum lead was 0.02 μg/dL. The chest radiograph showed a normal cardiac silhouette and normal pulmonary vascular markings. An ECG showed normal sinus rhythm with an inverted T wave in lead III. A Holter recording documented sinus rhythm, with rates ranging from 80 to 166 bpm and averaging 108 bpm. The hourly trend is shown in Fig 2⇓. Tachycardia occurred in the afternoon, from the hours of about 2 pm until midnight, corresponding to her reported symptoms. The average afternoon heart rate was 130 bpm, compared with 85 bpm at night. Pulmonary function tests showed normal lung volumes, mechanics, and diffusing capacity. A transthoracic echocardiogram with bubble contrast was normal (Fig 3⇓).
On a treadmill, she exercised for 6:57 minutes to stage II of the Bruce protocol with satisfactory effort; the study was terminated because of chest tightness. The heart rate rose from 121 bpm at rest to a maximum of 192 bpm at peak exercise. During recovery, her heart rate gradually fell to 140 bpm, where it stabilized for about 8 minutes. A Valsalva maneuver resulted in a pause, followed by a sudden decrease in heart rate to 121 bpm, and then a gradual reduction to normal resting rates.
Her symptoms progressed to the degree that she could no longer attend school, and she was tutored at home. She was admitted to the hospital for experienced observation with telemetry monitoring; however, no typical spells occurred during those 3 days. The ECG showed normal sinus rhythm (79 bpm), with similar T-wave axis and marginal voltage criteria for left ventricular hypertrophy. Serum calcium was 8.4 mg/dL; magnesium was 2.0 mg/dL. Liver and renal function tests were normal. A urine screen for heavy metals (antimony, arsenic, bismuth, and mercury) was negative. Creatine kinase was 55 U/L. Aldolase and cortisol levels were normal. A neurological consultant observed mild proximal muscle weakness, but such weakness was not present on subsequent examinations.
She was discharged with a transtelephonic event monitor that documented numerous episodes of sinus tachycardia of variable rate during typical spells. Cardiac catheterization and electrophysiology studies were recommended but were declined by the family. In response to family pressure for a less invasive evaluation, a TEE was offered. During the procedure, the patient gradually became tachycardic, with a heart rate of 140 to 178 bpm. She was noted to be extremely pale and slate gray in color. Systemic oxygen saturation was 82% by pulse oximetry. She became lethargic, weak, and unresponsive to painful stimuli. Her blood pressure was 110/75 mm Hg, and respiratory rate was 28 breaths per minute. Her cardiopulmonary examination was unchanged except for tachycardia, tachypnea, and cyanosis. There was no improvement in oxygenation or clinical status with administration of naloxone and 100% oxygen by face mask. Chest radiograph was normal, showing no evidence of infiltration, pneumothorax, or aspiration. A Holter monitor recording during the procedure showed sinus tachycardia. With great effort, the patient opened her eyes and slowly said, “Doctor, this feels just like one of my other spells … only a whole lot worse.”
George Schroth, MD
A transthoracic echocardiogram was performed with contrast. Two-dimensional imaging in the apical four-chamber view showed an apparent discontinuity of the interatrial septum (Fig 3⇑). Approximately 10 mL of agitated saline was injected into the right antecubital vein as the atria were imaged. No premature entry of contrast into the left atrium was demonstrated, and no negative contrast was observed in the right atrium.
For the TEE, the patient was premedicated with midazolam, meperidine, and promethazine, along with a topical anesthetic throat spray. This “cocktail” sedates the patient well enough to tolerate the procedure comfortably without full anesthesia. This was a good-quality study demonstrating normal chamber dimensions, wall motion, and valve function. The main area of interest was the interatrial septum, the continuity of which was confirmed. There was no evidence of left-to-right or right-to-left shunt by either bubble contrast or color flow Doppler.
Anne Hamilton Dougherty, MD
We are presented here with an interesting child whose history is filled with tantalizing hints of pathology, many of which may be red herrings. We begin with an adolescent who has recurrent spells, always in the afternoon, characterized by palpitations, dyspnea, headache, lethargy, myalgia, and pallor. She was the product of an emergency preterm delivery and required oxygen and ventilation as a neonate. Her development had been very delayed. She suffered from mild asthma and environmental allergies. There was a notable history of exercise avoidance and incapacitation; this led to school avoidance and possible secondary gain. She also had chronic exposure to rice fields (a source of fertilizers, pesticides, molds, and other allergens) and to a complex mixture of toxic waste and well water. Her parents also have a source of possible secondary gain from her illness because of pending litigation. Any objective physician must at least consider the possibility of a functional disorder on the part of the patient or even a Munchausen’s syndrome by proxy on the part of her family.
At times, she had a resting sinus tachycardia, exacerbated by exertion. Either by design or desire, this may have been a consequence of her poor physical conditioning. The murmur described is consistent with a Still’s murmur, one of the most common innocent murmurs of childhood. Typically, one hears a short vibratory systolic murmur at the left lower sternal border and apex that is accentuated in the supine position or with exertion and fever. Although she had a normal S2, one of the three echocardiograms showed echo “dropout” in the interatrial septum with a left-to-right shunt. The plane of the atrial septum lies parallel to the path of the ultrasonic beam in the parasternal and apical views producing the apparent defect, which is a false-positive finding, possibly reflecting inexperience of the technician and/or the interpreter. The subxiphoid and posterior views are more reliable. The accuracy of Doppler flow studies is also subject to technician experience.
Sinus tachycardia was documented during several of her spontaneous spells. She had normal blood count, heavy-metal screen, chest radiograph, and pulmonary function tests and a nonspecific ECG. There was no clubbing or polycythemia to suggest chronic cyanosis. Thyroid function tests were normal, although it was not specifically stated that a T3 radioimmunoassay was performed to rule out an isolated T3 toxicosis. Finally, a TEE was complicated by the acute development of tachycardia and slate-gray cyanosis that did not respond to oxygen administration.
Accounting for “spells” is one of the most difficult tasks we attempt, because events tend to occur randomly, punctuating long periods of normality. My diagnostic strategy is to (1) document a spontaneous symptomatic episode (using Holter monitoring, telemetry, a transient symptomatic event recorder, or occasionally an electroencephalogram), (2) provoke an episode reproducing the clinical symptoms (using a treadmill test, electrophysiology study, or tilt-table test), or (3) attempt to suppress symptoms with a trial of empiric treatment, usually an antiarrhythmic or anticonvulsant agent. In this case, the child received trials of both propranolol and verapamil, neither of which helped.
Her best-described initial symptom was palpitation. A patient may perceive palpitations when the heartbeat is too fast, irregular, or more forceful than usual. The beat may be more noticeable in certain positions in which the heart is close to the chest wall. Rarely, bradycardia causes palpitations.
The symptom history is critical. In this case, it may have been misleading because of the high level of family anxiety and their relatively unsophisticated background. The consistent occurrence of tachycardia in the afternoon and the prolonged incapacitation are intriguing features to which we will return later. All of the associated symptoms are nonspecific until the final event. She was described as “like a statue,” which I would have interpreted as pale and marbled in complexion, unresponsive, or catatonic. Positional changes associated with palpitations may be helpful. For instance, tachycardia that occurs on standing is consistent with orthostatic hypotension. Spells that occur with sitting or bending suggest an atrial mass with intermittent obstruction of an atrioventricular valve. Spells that are relieved with squatting, especially those associated with hyperpnea and cyanosis, suggest tetralogy of Fallot. An abrupt onset and offset of symptoms suggests an arrhythmia with a reentrant mechanism, as does a response to vagal maneuvers. Precipitating factors, such as exercise, stress, and emotion, may also be important clues that suggest potentially helpful provocative maneuvers to the physician. A history of the use of drugs or stimulants (eg, caffeine, cocaine, decongestants, bronchodilators) suggests causality if there is a temporal relation between exposure and symptoms. Symptoms that appear after the event may also be helpful, eg, diuresis following an episode of PSVT. Could her prolonged lethargy constitute a postictal state?
Palpitations in this child were associated with sinus tachycardia, and several typical spells were completely documented electrocardiographically from onset to offset. Tachycardia may by physiological, pharmacological, or pathological. Physiological triggers include exercise, emotion, fever, hemodynamic or respiratory compromise, anemia, and poor physical conditioning. By design or desire, regular exercise was not a part of this patient’s routine; however, the tachycardia of poor physical conditioning is accentuated with effort and recedes with rest. It does not continue for hours at rest, as in this child. The hemodynamic response to a fall in arterial blood pressure or dehydration is a compensatory sinus tachycardia, as it is to intermittent bronchospasm or arterial desaturation.
Pharmacological causes include all sympathomimetic and vagolytic agents. Could she have received any of these? Patients may not always volunteer information about the use of nonprescription drugs, such as caffeine and decongestants, so a specific interrogation as to the use of any these agents is helpful. Could she have surreptitiously received someone else’s medication? We are told that her mother took propranolol and digoxin which, if anything, should have had the opposite effect.
Among pathological conditions that cause a secondary sinus tachycardia, the most common is thyrotoxicosis. Could she have an isolated T3 toxicosis? The intermittency of symptoms seems a poor fit. Pheochromocytomas, on the other hand, may intermittently release catecholamines, resulting in paroxysmal spells with labile hypertension, tachyarrhythmias, diaphoresis, and orthostatic hypotension. Affected patients may be particularly susceptible to spells during medical procedures. Altered parasympathetic innervation of the heart is a rare condition resulting in spells of inappropriate tachycardia. It has been described after radiofrequency catheter ablative therapy, which our patient did not receive. Sinus node reentry tachycardia, a relatively rare form of PSVT, may imitate sinus tachycardia. The P-wave morphology is similar or even identical to that of sinus rhythms. The onset and offset, however, are abrupt rather than gradual, and tachycardia can be reproduced with programmed stimulation of the atrium. Although the patient did have a plateau in heart rate while recovering from treadmill exercise, the gradual acceleration to peak exercise, gradual deceleration to 140 bpm in recovery, and gradual fall after Valsalva would be more consistent with sinus tachycardia than with sinus node reentry. Finally, DaCosta’s syndrome, or posttraumatic stress syndrome, may cause inappropriate sinus tachycardia.
This patient experienced an unusual complication of TEE: profound cyanosis! Most important, the episode reproduced her clinical symptoms. At this juncture, one can effectively rule out functional disorders.
The causes of central cyanosis are relatively few and can be grouped into conditions with decreased oxygen available to hemoglobin and those with an abnormal hemoglobin that prevents oxygen transport. Arterial oxygen tension may fall as a consequence of pulmonary disease, hypoventilation, or a right-to-left cardiac shunt. Sudden or paroxysmal pulmonary causes include pulmonary emboli, asthma, and pneumothorax. Recurrent emboli are most unusual in childhood, unless there is a hypercoagulable state. Furthermore, the timing of the event during rather than after TEE seems atypical, as does the occurrence of chronic paroxysmal symptoms only in the afternoon. Hypoventilation secondary to oversedation is a consideration; however, her respiratory rate was 28 breaths per minute, and the cyanosis did not reverse with naloxone. Obstructive sleep apnea does not quite fit the diurnal pattern we see in this case.
What about the possibility of cyanotic heart disease? One would expect a resting cyanosis in most cases, although it might be exacerbated with effort. One would also expect other stigmata of a chronic cyanotic condition, such as polycythemia or clubbing. The possibility of an atrial septal defect has been offered. Normally associated with a left-to-right shunt, a long-standing significant septal defect can result in irreversible pulmonary hypertension and, thus, reverse the shunt; this is the Eisenmenger reaction. This possibility, however, was effectively ruled out with echocardiography. Of all the common right-to-left shunts, only a pulmonary arteriovenous fistula might be overlooked on a high-quality echocardiogram. That condition would be chronic, and one would expect to hear a continuous murmur and likely see evidence of Rendu-Osler-Weber syndrome as well.
We are left with abnormal hemoglobins to consider. Methemoglobin is the form of hemoglobin in which the iron molecule is oxidized by electron transfer rather than by association with oxygen. This reaction creates a stable ferric state that is incapable of binding oxygen for transport. The small amount of methemoglobin that normally occurs in erythrocytes is reduced effectively by an NADH-dependent enzyme, cytochrome b5 reductase (methemoglobin reductase, diaphorase). Methemoglobinemia is a rare condition, either inherited or acquired, in which this enzyme capability is either reduced or overwhelmed. Clinical cyanosis occurs with total methemoglobin levels of ≈1.5 g/dL or greater, depending on the darkness of the skin. The arterial blood is a characteristic chocolate or mahogany brown color and does not pink up, even when mixed with air. Methemoglobinemia would account for this patient’s persistent cyanosis when treated with 100% oxygen.
Hereditary methemoglobinemia results from either abnormal M hemoglobins or from a deficiency in cytochrome b5 reductase. M hemoglobins are transmitted in an autosomal-dominant fashion and result in cyanosis from birth with minimal symptoms. Reductase deficiency, on the other hand, is an autosomal-recessive condition with a predilection for Inuit and Alaskan Native Americans and for certain families in Appalachia. Homozygotes have chronic cyanosis, whereas heterozygotes are usually asymptomatic but may be predisposed to cyanosis with oxidant stress. Newborn heterozygotes may be cyanotic.
Acquired methemoglobinemia results from toxic exposure to oxidants and does not require an underlying reductase deficiency. Nitrites and nitrates, local anesthetics, and acetaminophen are among the most common offending medicinals. Berton Roueché’s classic tale of medical detective work, Eleven Blue Men, describes the accidental poisoning of New York flophouse residents with sodium nitrate–tainted oatmeal.1 Hundreds of chemicals have been reported to cause the condition.2 Many are found in common household products, such as aniline dyes, disinfectants, mothballs, menthol, herbicides, fungicides, perfumes, and synthetic rubber. Ingestion of well water contaminated by nitrogen from fertilizers or fecal matter is a well-established cause of methemoglobinemia in susceptible individuals, especially infants. This association was initially described by an observant pediatric resident in 1945.3 Neonates are particularly susceptible because of a relative deficiency in cytochrome b5 reductase activity before the age of 6 months.4 There are case reports of the occurrence of methemoglobinemia due to the use of local anesthetics during esophagogastroduodenoscopy and bronchoscopy.5 Our patient would have routinely received topical sprays containing lidocaine and benzocaine, as well as lidocaine jelly lubricant during her TEE. This complication has been reported in individuals with normal reductase levels and normal mucosal barriers with routine doses of anesthetic, suggesting that some individuals are simply more susceptible than others.
Congenital hemoglobin variants with low oxygen affinity are quite rare and result in persistent rather than paroxysmal cyanosis. Sulfhemoglobinemia is even more rare; in this entity, sulfur replaces the iron molecule in the porphyrin ring. It has been reported to be caused by phenacetin, acetaminophen, and hydrogen sulfide. There is also a rare congenital form. Symptoms tend to be mild, and treatment is usually unnecessary.
Methemoglobinemia is a plausible diagnosis in this case because it ties together most of the patient’s history and the final cyanotic event. Why her spells occurred in the afternoon remains unclear. The pallor described in her usual spells was probably a mild cyanosis, and the sinus tachycardia was probably secondary to intermittent desaturation. She had ample opportunity to come into contact with environmental oxidants, including well water, acetaminophen, and agricultural chemicals. Countless household and industrial waste products also may have triggered events. Separated from her normal environment, she had no spells during prolonged observation in the hospital. The culprits during the final procedure were most certainly lidocaine and benzocaine. Cytochrome b5 reductase deficiency may have predisposed her to oxidant sensitivity, but it is not a necessary risk factor for the condition.
During the TEE, her gradual sinus tachycardia was noted before the cyanosis was detected. Of note, the noninvasive pulse oximeters routinely used to monitor oxygen saturation during endoscopy procedures may be insufficient to detect developing methemoglobinemia. These devices measure red (660-nm) and infrared (940-nm) light transmission through intact erythrocytes in tissue. Methemoglobin absorbs at both of these wavelengths, as does oxyhemoglobin. In high concentrations, the ratio of absorption at these two wavelengths confuses the calibration curve and may force the measured saturation to 85%, independent of actual oxygen saturation.6
This patient received oxygen and naloxone without improving. On reexamination, she had no new crackles, wheezes, jugular venous distension, or murmurs and no evidence of pneumothorax. I think that an arterial blood gas will lead us in the right direction. If, as I suspect, the measured Sao2 is low with a normal or high Pao2 then administration of intravenous methylene blue would be the treatment of choice. Methemoglobin and cytochrome b5 reductase levels and hemoglobin electrophoresis will confirm the diagnosis.
Roma L. Ilkiw, MD
Indeed, Dr Dougherty is correct in her diagnosis. Methemoglobinemia is suspected when the measured Sao2 is reduced, despite a normal Pao2. Arterial blood was drawn while the patient was breathing 100% oxygen, revealing a pH of 7.37, Pao2 of 316 mm Hg, Paco2 of 45 mm Hg, HCO3 of 26 mmol/L, and base excess of 1 mmol/L. Oxygen saturation was calculated from the Pao2 at 99%, in contrast to the oximetric measurement of 80%. A methemoglobin level was 16.9% 4 hours after the procedure and 5.5% at 8 hours, with a gradual decrease to normal. We speculate that the patient’s methemoglobin levels were higher at the time of maximal symptoms; however, there is no way to document that retrospectively. The cyanosis of methemoglobinemia is appreciated when methemoglobin levels exceed 10%. Symptoms of irritability, lethargy, and exercise intolerance may occur with methemoglobin levels of 30% to 40%. Levels >50% are associated with central nervous system depression, and levels >70% are incompatible with life. This patient’s methemoglobin reductase level was 13.5 IU/g hemoglobin (normal, 10.1 to 19.4 IU/g). Hemoglobin electrophoresis showed a normal pattern. We concluded that the patient had an acquired methemoglobinemia consequent to environmental exposures as well as topical anesthetics applied during the TEE.
A medical toxicologist suggested that there was a significant potential for environmental exposure to agents causing methemoglobinemia in the patient’s home. He recommended inspection of the home by an industrial hygienist. The family, however, has moved approximately 30 miles away, and the patient has shown a remarkable resolution of her tachycardia.
Selected Abbreviations and Acronyms
|bpm||=||beats per minute|
|PSVT||=||paroxysmal supraventricular tachycardia|
This clinicopathological conference was presented at the University of Texas Medical School at Houston on March 27, 1995.
- Copyright © 1995 by American Heart Association