EARLY POSTOPERATIVE HEART FAILURE

Betta Medini and Alexandre Mebazaa, Département d’Anesthésie-Réanimation, Hôpital Lariboisière, Paris, France

Correspondence to: Alexandre Mebazaa, MD, PhD, Département d’Anesthésie-Réanimation, Hôpital Lariboisière, 2 rue A Paré, 75475 Paris Cedex 10, France. E-mail: alexandre.mebazaa@lrb.ap-hop-paris.fr

PRESENTATION AND HISTORY

An 89-year-old Caucasian male was admitted for elective total knee replacement surgery. His past medical history included treatment for essential hypertension and supraventricular arrhythmia since he was 69. At the age of 86, he had a posterior myocardial infarction with inferobasal akinesia, resulting in New York Heart Association (NYHA) II functional status. He had no history of cerebrovascular accidents, and cardiopulmonary examination was normal. Haemodynamic and blood analysis results are shown in Table 1. The preoperative electrocardiogram revealed sinus rhythm with numerous premature atrial contractions, QS waves in leads II, III, and a VF, in addition to a left anterior hemiblock (Figure 1).

The preoperative chest X-ray showed cardiac enlargement and apical emphysema (Figure 2). Echocardiography showed mild left ventricular dilatation (left ventricular end-diastolic diameter 55 mm) with slight left ventricular hypertrophy and inferobasal akinesia with preserved left ventricular function: the left ventricular ejection fraction was 44% (normal range 65%-85%).

Systolic pulmonary arterial pressure was approximately 30 mmHg. There was mild aortic regurgitation. Doppler of the carotid arteries showed a bilateral stenosis (40% on the left, 60% on the right). The American Society of Anesthesiologists (ASA) physical status was evaluated as III, and the Goldman Multifactorial Cardiac Risk Index¹ was assessed as 11/53. The patient was receiving treatment with lorazepam, aspirin, amiodarone, amlodipine, nitrates, and an association of captopril plus hydrochlorothiazide. Furthermore, he was being treated with beta-blocking ophthalmic drops for bilateral glaucoma.

All drugs were continued up to surgery except aspirin, which was replaced by flurbiprofen eight days before surgery, and captopril, which was replaced by clonidine 48 hours before surgery.
Low molecular weight heparin was administered the evening before the intervention, and oral hydroxyzine
(100 mg) and cimetidine were given one hour before the patient was taken to the operating theatre.

Back to Top

IN THE OPERATING THEATRE

Intraoperative monitoring included the electrocardiogram, pulse oximeter, invasive blood pressure, and transoesophageal Doppler. Before anaesthesia, the blood pressure was 160/90 mmHg and the heart rate 85 bpm. Oxygen saturation was 96% with room air. General anaesthesia was induced with sufentanil, etomidate, and atracurium, and maintained with boluses of sufentanil and isoflurane in a 50% mixture of oxygen and nitrous oxide. Oxygen saturation settled at around 96%.

Volume loading was performed with 2500 ml of crystalloids plus 1000 ml of colloids over four hours, because the patient was considered dehydrated before surgery; intraoperative bleeding was estimated to be 1000 ml. The initial postoperative haemoglobin was 10.4 g/dl. During the intervention there was no urine output. Throughout the intervention, systolic blood pressure remained stable between 100 and 120 mmHg, diastolic blood pressure between 50 and 60 mmHg, and heart rate between 50 and 60 bpm. Only a few premature atrial contractions were seen during anaesthesia. Stroke volume was 51 ml at incision and remained stable between 50 and 60 ml during the intervention (normal range 60-90 ml).

Back to Top

IN THE RECOVERY ROOM

Hour 0 (H0) Half an hour after admission to the recovery room, the patient was extubated and oxygen saturation was 98% on 6 litres/minute of oxygen via face mask. Blood pressure was 130/60 mmHg, heart rate 70 bpm, intratympanic temperature 34.8°C, and the patient was shivering. The first postoperative electrocardiogram showed an increased number of premature atrial contractions and a premature ventricular contraction, without new ST segment abnormalities (Figure 3). The patient received paracetamol and nefopam according to the acute pain control protocol and his pain score remained lower than 2/10 during the entire postoperative period.

Although volume loading was continued with 500 ml of crystalloids and 500 ml of colloids, the patient remained anuric. Three hours later, diuresis (1000 ml) was induced by an intravenous dose of furosemide (40 mg), which also induced haemoconcentration (proteins rose from 49 to 52 g/l), hypokalaemia (3.4 mmol/l), and hypocalcaemia (1.80 mmol/l).

H7 The electrocardiogram showed the first signs of myocardial ischaemia: increased amplitude of the T wave and the presence of a 1-mm elevation in the ST segment in the precordial leads (V3 through V6) (Figure 3). Troponin increased from 0.6 to 2.5 µg/l. At this time, the blood pressure was 130/70 mmHg and the heart rate had increased to 92 bpm. It appeared that myocardial ischaemia may have been related to the decrease in haemoglobin (from 10.4 to 8.9 g/dl) due to significant postoperative bleeding (1400 ml). Thus, the patient received three units of red blood cells over six hours while bleeding in the drains stopped.

H19 Despite an improvement in the haemoglobin level (10.8 g/dl), adequate fluid management, and correction of the electrolyte imbalance (Table 1), the patient became more tachycardic (heart rate 114 bpm) and started to complain of dyspnoea. The blood pressure was 140/70 mmHg. The patient did not complain of precordial pain but became tachypnoeic (25 breaths/minute) and hypoxaemic despite an increase in supplemental oxygen flow rate from two to six litres per minute. Arterial blood gas measurements were: pH 7.50, partial oxygen pressure 62 mmHg, partial carbon dioxide pressure 31mmHg, and oxygen saturation 93%.

Pulmonary auscultation revealed bilateral rales and the chest X-ray showed diffuse haziness of the lung fields, predominantly in the hilar regions, and loss of a distinct vascular margin (Figure 2).

A fourth heart sound was audible on cardiac auscultation. The corresponding electrocardiogram measurement showed ST segment elevation in leads III and V3 through V6 (Figure 3). Troponin was 14.7 µg/l.

Echocardiography demonstrated additional apical akinesia compared with preoperative examination. Nevertheless, the left ventricular ejection fraction was 42% and stable compared with the preoperative value. The right ventricle was normokinetic. Systolic pulmonary artery pressure was 40 mmHg.

Back to Top

DIAGNOSIS

This patient showed early postoperative myocardial ischaemia due to decompensated coronary artery disease. This induced pulmonary congestion, probably related to left ventricular diastolic failure, while left ventricular systolic function was preserved.

Back to Top

MANAGEMENT AND TREATMENT

Nitrates and continuous positive airway pressure were immediately started to improve coronary oxygen delivery. In addition, esmolol, a short-acting beta-blocker, was administered intravenously at an infusion rate of 50 mg/hour in an attempt to reduce the heart rate and, subsequently, myocardial oxygen consumption. Esmolol administration slowed the heart rate to 100 bpm. At H31, the effect of the esmolol was striking, with improvements in clinical, electrocardiographic, and blood gas variables; troponin values peaked at 143.4 µg/l before returning toward normal values. The esmolol was substituted by atenolol, a longer-acting beta-blocker, which was given orally at a dose of 100 mg/day.

The patient was discharged from the recovery room at H48. A coronary angiography performed one week later showed multiple lesions in all three coronary arteries.

Back to Top

LESSONS TO BE LEARNED

Myocardial ischaemia has always been a frequent perioperative cardiac complication,² and the effect of the ageing of the general population over the last decade has exacerbated this problem. Myocardial ischaemia mainly occurs during the early postoperative phase. The peak incidence of perioperative myocardial infarction (the most severe consequence of myocardial ischaemia) occurs within the first three days after the operation³ and, more precisely, between 12 and 32 hours after surgery; it is always preceded by myocardial ischaemic events.^4,5 In addition, the survival rate two years after noncardiac surgery is reduced in patients with coronary artery disease compared with control patients (79% vs. 93%).⁶ Postoperative decompensation of coronary artery disease is usually silent, even in nondiabetic patients,⁷ and is often detected only when myocardial ischaemia is advanced and ventricular failure becomes manifest.

In our patient, myocardial ischaemia was detected in the electrocardiogram performed seven hours after surgery (H7), when troponin was slightly increased (2.5 µg/l) and the patient showed no clinical signs except a small increase in heart rate (92 bpm). At that time, myocardial ischaemia was related to an imbalance in myocardial oxygen supply and demand. On one hand, the total oxygen demand of the body, and especially myocardial oxygen demand, was increased by the rise in body temperature from 34.5°C to 36°C over 30 minutes, the patient’s shivering, and the subsequent increase in heart rate. On the other hand, despite an adequate left coronary perfusion pressure (diastolic blood pressure >65 mmHg) and an arterial oxygen saturation greater than 97%, myocardial oxygen supply was inadequate, with a haemoglobin level <9 g/dl in a patient at high risk of perioperative cardiac complications (age over 70 years, prior myocardial infarction, limited exercise capacity⁸). In addition, the haemoglobin level was probably overestimated because of the furosemide-induced haemoconcentration.

Over the following 12 hours, despite transfusion of three units of red blood cells, the heart rate continued to increase because of the persistence of myocardial ischaemia in association with electrolyte disorders (hypokalaemia and hypocalcaemia^9,10). Indeed, tachycardia further increased myocardial oxygen demand and decreased myocardial oxygen supply by reducing left coronary diastolic filling time. Accordingly, as shown in Figure 4, a vicious cycle developed, with tachycardia worsening myocardial ischaemia and inducing left ventricular dysfunction.

Ventricular failure became manifest with pulmonary oedema 19 hours into the postoperative period. At that time, there were no signs of measurable systolic ventricular failure: the left ventricular ejection fraction remained stable before and after pulmonary congestion, the blood pressure remained stable, and no signs of ischaemia could be detected in the major organs (brain, liver, and kidneys). Thus, the pulmonary congestion was probably related to an exacerbation of diastolic dysfunction due to myocardial ischaemia.

Analysis of cardiac physiology can explain why patients with coronary artery disease, and particularly those with left ventricular hypertrophy, are susceptible to diastolic heart failure. When haemodynamically challenged by stress, such as tachycardia, persons with coronary artery disease and left ventricular hypertrophy are unable to supply adequate oxygen to the left ventricular wall. Tachycardia and the subsequent myocardial ischaemia may therefore worsen left ventricular relaxation and compliance, which are both already reduced by left ventricular hypertrophy.¹¹ Consequently a cascade of events begins, in which the left ventricular end-diastolic pressure rises, left atrial pressure increases, and pulmonary oedema develops.¹²

The treatment of patients with left ventricular diastolic dysfunction remains empirical. Current treatment includes avoiding excessive sodium intake, cautious use of diuretics (to relieve pulmonary congestion without excessive reduction in preload), restoration and maintenance of sinus rhythm at a heart rate that optimises ventricular filling, and correction of precipitating factors such as acute ischaemia and elevated blood pressure.¹³

There is enthusiasm for treating left ventricular diastolic dysfunction with various classes of medications such as calcium channel blockers and ACE inhibitors, because they seem to improve surrogate outcome measures. However, few clinical data currently exist to support the efficacy of any particular class of drugs for left ventricular diastolic dysfunction. In our patient we chose to treat the tachycardia, which was the major cause of the myocardial ischaemia and the subsequent left ventricular dysfunction.¹⁴ Since left ventricular systolic function appeared to remain stable with a stable blood pressure, we decided to administer esmolol, an intravenous, short-acting beta-blocker. This could be seen as a paradoxical treatment for pulmonary oedema but, in the presence of stable systolic function, esmolol was aimed at reducing heart rate and myocardial ischaemia. This proved successful, as the patient ultimately recovered from the pulmonary congestion.

Back to Top

A CLINICAL TRIAL RELATED TO THE CASE

A recent paper, from Boersma et al,⁸ published in the Journal of the American Medical Association, came from a team who worked for several years on the assessment of ‘predictors of cardiac events after major surgery’ in patients with potential coronary artery disease. These authors included 1351 consecutive patients scheduled for elective major vascular surgery, 72% of whom had at least one of the following cardiovascular risk factors: age >70 years, angina, prior myocardial infarction, congestive heart failure, treatment for ventricular arrhythmias, treatment for diabetes mellitus, limited exercise capacity, hyperlipidaemia, or smoking. They found 45 perioperative cardiac complications (3.3% of all patients): 31 patients had cardiac death and another 14 nonfatal myocardial infarction. Although these authors could not assess the number of patients in whom an episode of myocardial ischaemia occurred without death or myocardial infarction, it is usually assumed that the number of ischaemic episodes is three to five times greater than the number of deaths and myocardial infarctions.

In this study, the authors demonstrated that advanced age, current or prior angina, and a history of cardiac or cerebral events are the most important clinical determinants of perioperative cardiac death or myocardial infarction. They also showed that dobutamine stress echocardiography results have a good prognostic value for adverse cardiac outcome in patients with more than three risk factors.

This paper also evaluated the effects of preoperative beta-blocker therapy on the occurrence of postoperative cardiac events. There are, indeed, increasing data on the prophylactic efficacy of beta-blockers in reducing perioperative myocardial ischaemia and improving long-term survival after major surgery.^15,16,17 In their paper, Boersma et al⁸ showed that, regardless of the number of clinical risk factors, patients receiving beta-blockers (n = 360) had a significantly lower risk of cardiac events than those who did not receive them. The cardiac complication rate in patients without risk factors was 1.2% in patients who were not beta-blocker users and 0% in those who were. More importantly, in patients with one to three risk factors, the cardiac complication rate was 3.0% in patients who did not receive beta-blockers and 0.9% in those who did. Finally, in patients with more than three risk factors, the cardiac complication rate was 12.6% in patients not using beta-blockers and 5.9% in beta-blocker users. In addition, it was emphasized that patients receiving beta-blockers had a considerably worse overall profile than those who did not receive them, which makes this result even more convincing.

In the current case, despite the existence of more than three clinical risk factors for cardiac complications, the patient did not receive prophylactic beta-blockers because the anaesthetists and cardiologists expected this treatment to worsen systolic function. However, since postoperative tachycardia-induced myocardial ischaemia occurred, and we successfully treated this with a beta-blocker, it appears that beta-blockers could have been given in the preoperative period.

In summary, the best therapy for perioperative cardiac complications is to prevent them, by avoiding precipitating factors and by giving appropriate prophylactic therapy.

Back to Top

©2002 Failinghearts.com