Delayed-onset Epinephrine-induced Pulmonary Edema 

A 25-yr-old, 70-kg woman was scheduled for wound debridement. Seventeen days previously she was injured, resulting in a left femoral segmental open fracture and a tibia–fibular fracture. Before this operation, she received general anesthesia three times for open reduction and internal fixation, debridement, and sequestrectomy. She was scheduled for debridement of three areas (a 10-× 5-cm open wound, a 20-× 4-cm open wound, and a 3-cm laceration wound) of her left leg.

Preanesthetic blood pressure (BP) was 131/77 mmHg and heart rate was 115 beats/min. One day before surgery, laboratory data showed the following: hemoglobin, 9.2 g/dl; hematocrit, 27.6%; blood urea nitrogen, 18 mg/dl; creatinine, 1.8 mg/dl; potassium, 4.2 mmol/l. A central venous pressure line was inserted via  the right femoral vein, and 500 ml packed erythrocytes were transfused for correction of the anemia.

Standard noninvasive monitors were placed. Anesthesia was induced with 100 μg fentanyl and 300 mg thiopental, with 35 mg atracurium given to facilitate tracheal intubation. Anesthesia was maintained with isoflurane, nitrous oxide, and oxygen. The anesthesia course was uneventful for approximately 3 h. Total blood loss was 500 ml; the amount of intraoperative fluid administration (D51/4NS plus Ringer's Lactate) was 1,600 ml, and urine output was 1,200 ml. However, at the end of surgery, several pieces of gauze dressing containing diluted epinephrine (epinephrine solution, 0.1 w/v%, 100 ml per bottle; DAIICHI Pharmaceutical Co., Ltd., Tokyo, Japan) were placed over the wound area. The gauze dressings were said to contain a total amount of 100 ml epinephrine solution, 1:100,000. Hypertension (BP near 180/100 mmHg) and tachycardia (120 beats/min) with occasional ventricular premature contractions were noted immediately after the gauze placement. The patient was awakened and extubated in the operation room and was sent to the recovery room.

At arrival in the recovery room, the patient was conscious. BP was 165/115 mmHg; heart rate was 110 beats/min, and respiratory rate was 20 breaths/min. Oxygen saturation by pulse oximetry (Sp^O2) was 99%. One hour later, BP decreased to 106/74 mmHg and tachycardia (134–140 beats/min) was noted. Arterial blood showed the following:p  H, 7.35; arterial oxygen tension (Pa^O2), 64 mmHg; arterial carbon dioxide tension (Pa^CO2), 34 mmHg; bicarbonate (HCO³⁻), 19 mmol/l; base excess, −5.4; hemoglobin, 8.9 g/dl, arterial oxygen saturation (Sa^O2), 91.6%; sodium, 131 mM/l; and potassium, 2.26 mM/l. Five hundred milliliters packed erythrocytes and 20 mEq KCl in 800 ml Ringer's lactate were infused slowly. Heart rate increased rapidly to 145 beats/min despite treatment, chest auscultation revealed bilateral moist rales, and central venous pressure was 27 cm H²O with poor urine output. Twenty milligrams furosemide was given. Chest radiography, performed 50 min later, confirmed the diagnosis of pulmonary edema. One hour later, hypotension (82/59 mmHg) developed. The patient became hypoxemic (Sp^O2< 85%) and began coughing up frothy pink sputum. Tracheal intubation was immediately performed after 10 mg diazepam and 80 mg succinylcholine was given intravenously.

Polymorphic ventricular tachycardia with no palpable cardiac output was noted after intubation. Cardiopulmonary resuscitation (CPR) was immediately started. An arterial blood sample, drawn 20 min after intubation, revealed the following:p  H, 7.09; Pa^O2, 45 mmHg; Pa^CO+2, 52 mmHg; hemoglobin, 12.5 g/dl; BE, −14.7; Sa^O2, 63.7%; sodium, 131 mM/l; potassium, 8.0 mM/l; and bicarbonate, 15.6 mM/l. The patient was treated with epinephrine, direct current defibrillation, lidocaine, sodium bicarbonate, calcium chloride, and regular insulin. Potassium concentrations obtained at 45, 60, 70, and 120 min after intubation were 13.3, 6.3, 7.2, and 5.8 mmol/l, respectively. After 2 h 40 min of CPR, sinus rhythm reappeared with a heart rate of 130 beats/min and a BP of 112/58 mmHg. The patient regained consciousness later.

The patient was transferred to the intensive care unit and was treated with blood transfusion, intravenous furosemide, and hemodialysis because of post-CPR acute renal failure. A retrospective questioning of the scrub nurse revealed that 100 ml epinephrine, 1:1000 (100 mg), instead of 1:100,000 epinephrine had been inadvertently added into the pieces of gauze that were laid onto the surgical wound at the end of the surgery. Additional information from the ward showed that the patient had not been moving much in the previous 17 days because of nervousness and the fear of movement-associated pain.

During her stay in the intensive care unit there was no evidence of rhabdomyolysis. The patient was discharged to home without any neurologic sequelae after a 3-month hospital stay.

The use of epinephrine-soaked gauze is commonly used to stop bleeding. 5,6It typically results in some hypertension and tachycardia, but few other problems. 3,4,6In this case, however, a nursing error resulted in the use of perhaps a 10-fold increase in the dose of epinephrine to which the patient was exposed. Surprisingly, however, this did not result in a sudden deterioration in cardiovascular function, but rather led to the gradual development of pulmonary edema during a period of several hours after surgery. Typically pulmonary edema induced by epinephrine overdose occurs within 20 min of administration. 7However, in this patient, more than 1 h elapsed after placement of epinephrine-soaked gauze before the onset of pulmonary edema. The reason for this delay is not clear, but is presumably related to slow absorption of the applied drug through the damaged skin and a change that may have been related to the drug's local vasoconstrictive effect. Physicians should be aware that delayed complications from the topical administration of epinephrine may occur.

Unfortunately, our efforts to treat this patient's pulmonary edema and hypoxia resulted in another serious complication;i.e. , a hyperkalemic cardiac arrest, which was probably secondary to the administration of succinylcholine. Hyperkalemia after succinylcholine administration is well-described and can occur in patients with extensive trauma and prolonged immobilization. 8,9The use of succinylcholine was not considered a risk because the patient should have been able to move around after trauma despite her injuries. It was not until after resuscitation we knew that, because of nervousness and pain, she had not been moving much for the previous 17 days. In retrospect, severe hyperkalemia was perhaps predictable, and the use of succinylcholine in this patient was probably inappropriate. 10 

There have been articles that reported outcomes in prolonged resuscitation. 11,12This patient required almost 3 h of continuous closed chest CPR, and yet survived without neurologic sequelae. This supports the belief that continued resuscitative efforts in young persons may be worthwhile, particularly in the face of a potentially “reversible” or treatable cause, such as occurred in this patient.

In conclusion, we present a case of long-term immobilization that was complicated with inadvertent administration of high-dose epinephrine and succinylcholine-induced hyperkalemia cardiac arrest. Because absorption of epinephrine through the wound may be slow, late responses should be watched for carefully. Succinylcholine must be used with caution in patients with long-term disabilities, metabolic acidosis, and hypoxemia.