Pediatric Cardiology

, Volume 25, Issue 5, pp 558–561

Vasodilatory Shock After Surgery for Aortic Valve Endocarditis: Use of Low-Dose Vasopressin


  • E. Lechner
    • Department of Pediatric CardiologyChildren’s Heart Center Linz
  • H. A. Dickerson
    • Pediatric Cardiac Intensive Care Unit, Texas Children’s Hospital and Department of PediatricsBaylor College of Medicine
  • C. D. FraserJr.
    • Congenital Heart Surgery, Texas Children’s Hospital and Department of SurgeryBaylor College of Medicine
    • Pediatric Cardiac Intensive Care Unit, Texas Children’s Hospital and Department of PediatricsBaylor College of Medicine

DOI: 10.1007/s00246-003-0544-x

Cite this article as:
Lechner, E., Dickerson, H.A., Fraser, C.D. et al. Pediatr Cardiol (2004) 25: 558. doi:10.1007/s00246-003-0544-x


This is the case report of a 13-year-old male who developed vasopressor-resistant hypotension after cardiac surgery for endocarditis. As norepinephrine resulted in aggravation of the preexisting ventricular arrhythmia, vasopressin was used to maintain blood pressure. The vasopressin continous infusion was started at 0.00002 units/kg/min and titrated up to 0.0003 U/kg/min. This low dose led to resolution of hypotension without causing side effects. As the appropriate indication and dose of vasopressin is not established, the cautious use of vasopressin in children is recommended.


VasopressinVasodilatory shockEndocarditisCardiopulmonary bypass—children

Patients with staphylococcal endocarditis of the aortic or mitral valve are at high risk to develop complications such as severe congestive heart failure with unstable hemodynamic profile, perivalvular extension of the infection, or systemic embolization. When complications are recognized, treatment should be combined antibiotic therapy and surgery, as mortality rates among patients who have staphylococcal endocarditis involving the aortic valve and who were treated surgically range from 5% to 26% (compared with 51% of those treated with antibiotics alone [3, 5]).

Systemic vasodilation and severe hypotension can occur due to septic shock or from systemic inflammatory response after cardiopulmonary bypass. Vasopressin, a direct systemic vasoconstrictor and an antidiuretic hormone, has been used in high doses as a splanchnic vasoconstrictor in patients with variceal bleeding [12].

The successful therapy of vasodilatory shock secondary to sepsis or systemic inflammatory response syndrome after postcardiopulmonary bypass in adults with vasopressin has been reported previously [2, 4, 6, 8, 9, 10]. Data on the use of vasopressin in children with vasodilatory shock, however, are very limited, and indications and dosing have not been established [7, 10].

This case demonstrates the use of vasopressin in the treatment of vasodilatory shock after cardiac surgery in a child with Staphylococcus aureus endocarditis.

Case Report

A 13-year-old previously healthy male without a known cardiac history presented with persistent fever and new heart murmur. Seven days prior to admission he had fevers up to 40°C. He complained of abdominal discomfort with vomiting, weakness, myalgias, and headache. A rapid Strep test on the third day of illness was negative, but white cell count and platelets were decreased. He received cephalexin for 2 days. On day five of illness a blood culture was drawn. The mother reported a minor injury from a splinter on the patient’s right foot, which had occurred 3 weeks prior to the onset of fever. On admission, he had a rash on his back, scattered petechiae and splinter hemorrhages on his palms, soles, fingers, and toes. In addition, he had a slightly inflamed healing abrasion on his right foot. A new grade 3/6 systolic and grade 3/6 diastolic murmur at the left sternal border and apex were ascultated. The pulses were bounding, and he had a palpable liver and spleen. Heart rate was 90 bpm, blood pressure 129/55 mmHg, and oxygen saturation was 98%.

Initial laboratory data revealed a white blood cell count of 15,460 cells/μL (79% segmented, 8% bands), a platelet count of 55,000 cells μL, a C-reactive protein of 25.7 mg/dL, and a mild coagulopathy with mildly elevated liver functions tests.

Chest X-ray demonstrated small bilateral pleural effusions and mild cardiomegaly. An echocardiogram revealed a bicuspid aortic valve with thickened leaflets, no aortic stenosis, and moderate aortic regurgitation. The aortic root was thickened and echogenic, consistent with an aortic root abscess. No discrete vegetations were seen. The left ventricle had normal dimensions and a systolic ventricular shortening fraction of 40%. An MRI of the brain done as part of the workup revealed multiple scattered areas of gliotic change and gyral swelling consistent with cerebral emboli. An abdominal ultrasound and computer tomography demonstrated an inhomogeneous lesion at the inferior splenic margin consistent with a splenic infarct or abscess.

Three separate blood cultures later were positive for methicillin-susceptible Stapylococcus aureus. Antibiotic therapy was initiated with vancomycin and gentamicin. This regimen was switched to nafcillin, rifampin, and gentamicin when the results of the resistance profile of the staphylococcos were available. Three days later an echocardiogram noted a new mobile vegetation on the anterior mitral leaflet without mitral regurgitation.

The patient underwent aortic root homograft replacement, mitral valve debridement with repair of the mitral annulus and aortic root with autologous pericardium, and closure of a patent foramen ovale. There was a cloudy, purulent-appearing pericardial effusion, out of which few white blood cells, but no organisms, were isolated. The operative course was uneventful. He came off bypass on dopamine at 5 μg/kg/min and nitroprusside at 0.5 μg/kg/min. As he had runs of ventricular tachycardia while coming off bypass, lidocaine was started at 20 μg/kg/min and increased to 40 μg/kg/min. The postoperative transesophageal echocardiogram revealed mild aortic valve regurgitation, no significant mitral valve regurgitation, and good biventricular systolic function.

Four hours postoperatively, he became hypotensive (systolic blood pressure < 80 mmHg, mean blood pressure < 40 mmHg). A fentanyl drip was used for sedation without any changes in dosage during this period. The patient was paralyzed with vecuronium bolus; the last bolus was given 2 h before the hypotension occurred. Nitroprusside was discontinued, dopamine was increased to 10 μg/kg/min, 500 mg of CaCl and 250 mL of albumin were administered, and norepinephrine was started at 0.05 μg/kg/min.

Norepinephrine had a positive effect on blood pressure, but was discontinued secondary to ventricular ectopy. Vasopressin was initiated 5 hours postoperatively at a dose of 0.00002 U/kg/min, and this initial, very low dose had no effect on blood pressure. The first positive effect was seen at a dose of 0.00005 U/kg/min and vasopressin was titrated to a maximum rate of 0.0003 U/kg/min to reach a mean arterial blood pressure of > 50 mmHg. (Fig. 1). No change in heart rate or filling pressures was seen during the vasopressin infusion. The echocardiogram on the first postoperative morning revealed no changes, especially no aggravation of aortic valve regurgitation or worsening of ventricular function. Serum sodium level and urine output did not change, and no signs of peripheral ischemia were observed throughout this period. Vasopressin was decreased 12 hours after initiation and discontinued within 14 hours. The patient was extubated and weaned off dopamine and lidocaine by the second postoperative day. He had no further hemodynamic instability throughout his postoperative course.
Figure 1

Systolic (s), mean (m), and diastolic (d) systemic blood pressure before and after administration of DOPA (dopamine, 10 μg/kg/min), NOR (norepinephrine, 0.05 μg/kg/min), and vasopressin (U/kg/min).


Low-dose continous infusions of vasopressin have been used recently in several trials in adults with vasodilatory shock secondary to sepsis or cardiac surgery [1, 6, 8, 9, 11]. It led to resolution of hypotension and reduction or withdrawal of catecholamine support, and it improved urine output without causing hyponatremia in adults. Sepsis and postcardiopulmonary bypass- induced severe vasodilation are associated with inappropriately low vasopressin plasma levels, so this vasopressin deficiency may contribute to the hypotension of vasodilatory shock [1, 6, 8, 10].

There is little experience in the use of vasopressin in children with vasodilatory shock. Rosenzweig et al. reported a case series of 11 children with refractory hypotension after cardiac surgery. After initiation of vasopressin at a dose of 0.0003–0.002 U/kg/min, blood pressure increased significantlly (systolic blood pressure rose from 65 ± 14 mmHg to 87 ± 17 mmHg) in all 11 children, and pressor dosages were reduced. As 2 patients who had poor left ventricular function died despite a transient improvement of blood pressure, vasopressin is probably less appropriate for left ventricular dysfunction. Liedel et al. reported vasopressin therapy for refractory hypotension secondary to sepsis in 5 children with severe underlying diseases: 1 premature 630-gram baby with necrotizing enterocolitis, 1 newborn with congenital diaphragmatic hernia, and 3 children with oncologic diseases with chemotherapy. The dose of vasopressin used in these patients ranged from 0.001 to 0.008 U/kg/min (Table 1). In all patients, blood pressure increased immediately, and catecholamine support was partially withdrawn in 4 patients [7].
Table 1

Summary of published articles on the use of vasopressin to treat vasodilatory shock


Trial design



Dose (U/min)

Reported side effects

Landry DW et al. [6], 1997

Case series

Septic shock

10 adults



Argenziano M et al. [1], 1998

Case series

Vasodilation after CPB

40 adults



Morales DLS et al. [8], 2000

Case series

Vasodilation after LVAD implantation

50 adults


Limb ischemia with tissue necrosis and ischemic bowel

Patel BM et al. [9], 2002

Randomized controlled trial

Septic shock

13 adults



Ryan DW et al. [11], 2002

Case report

Septic shock

2 adults



Trials in children

Dose (U/kg/min)

Rosenzweig EB et al. [10], 1999

Case series

Vasodilatory shock after CPB

11 children



Liedel JL et al. [7], 2002

Case report

Vasodilatory shock

5 children


Peripherial ischemia without skin necrosis

Abbrevations: CPB, cardiopulmonary bypass; LVAD, left ventricular assist device; n, number of patients included.

No studies have been done to determine the appropriate dose of vasopressin in either adults or children. In adult studies and case reports the doses have ranged from 0.004 to 0.1 U/min (Table 1) [1, 6, 8, 9, 11]. Extrapolated from these adult data, the dose for children should range in between 0.00006 and 0.001 U/kg/min. In the two reported case series in children, vasopressin was used at a dose of 0.0003–0.008 U/kg/min for a mean of 71 ± 46 h. (Table 1) [7, 10]. In the 16 previously described children, no significant side effects were reported. Two patients had peripheral vasoconstriction without skin necrosis at a dose of > 0.007 U/kg/min [7]. In the case of life-threatening gastrointestinal bleeding, a high-dose vasopressin infusion was used in children. At a rate of > 0.01 U/kg/min, it was associated with fluid overload, electrolyte abnormalities, and hypertension [12].

In all previous published studies in adults and in children, vasopressin was added to the existing therapy with traditional pressor agents.

In this case, the patient had a systemic inflammatory response with severe hypotension that was probably secondary to underlying staphylococcal sepsis and extended cardiopulmonary bypass. Administration of volume and high-dose dopamine had no effect on blood pressure; in addition, initiation of norepinephrine resulted in aggravation of the preexisting ventricular arrhythmia. Therefore, vasopressin was used as pressor agent to maintain blood pressure.

As the appropriate dose of vasopressin is not known, the vasopressin continuous infusion in this case was started at the very low dose of 0.00002 U/kg/min and slowly titrated up to 0.0003 U/kg/min to avoid side effects. Our major concern was not to induce splanchnic vasoconstriction in a patient with a history of possible splenic infarction. Therefore the initial dose was chosen from extrapolated adult data, which are lower than the reported doses used in children. During the vasopressin infusion, even at a very low dose, the blood pressure increased steadily. The patient was closely monitored and there were no clinical obvious signs of cutaneous, renal, or mesenteric ischemia. Vasopressin is a very potent drug with major side effects, and as its appropriate dose is not defined yet, it should be initiated at a low dose and slowly titrated, adjusted to effect, and not exceed an infusion rate of > 0.007 U/kg/min.


In selected patients with vasodilatory shock after cardiac surgery, vasopressin is an ideal agent compared to traditional pressor agents. Since indications, dosing, and duration of intravenous vasopressin therapy have not been established, its cautious use in children is recommended.

Copyright information

© Springer-Verlag 2004