|
|
||||||||
Eur J Cardiothorac Surg 1999;15:55-60
© 1999 Elsevier Science NL
Department of Cardiothoracic Surgery, St. Luke's Hospital, Guardamangia, Malta
Received 23 June 1998; received in revised form 6 October 1998; accepted 6 October 1998.
Corresponding author. Tel.: +356-2595-1738; fax: +356-240-176.
| Abstract |
|---|
|
|
|---|
Key Words: Angiotensin converting enzyme inhibitors Coronary artery bypass surgery Renal dysfunction Left ventricular dysfunction Haemodynamic tolerance
| Introduction |
|---|
|
|
|---|
It has been our practice to administer ACE inhibitors to all patients with a history of myocardial infarction and impaired ventricular function following coronary revascularisation. Previously, treatment usually started on the 5th post-operative day when renal and myocardial recovery is usually complete. However with ever decreasing in-hospital stay we attempt to commence ACE inhibitors on the first post-operative day. Detailed monitoring of renal function is usually continued until 48 h after surgery in preparation for the patient's discharge on the 4th or 5th post-operative day. For this reason it is desirable to commence ACE inhibitor treatment at a time when the patient's clinical parameters are closely monitored.
An alternative strategy would involve the primary care physicians commencing ACE inhibitor treatment in a stepwise fashion after the patient's discharge. Unfortunately, only 30% of our patients were taking ACE inhibitors pre-operatively when these were indicated. Moreover, medical therapy of many patients is not altered after discharge from hospital.
This study describes the degree of haemodynamic tolerance to two ACE inhibitors early after cardiac surgery with the purpose of assessing the feasability of discharging patients on such treatment, even when hospital stay is short.
| Materials and methods |
|---|
|
|
|---|
Approval for this research was granted by the Malta Medical Council and the Faculty of Medicine and Surgery Research Ethics Sub-Committee.
Patients were managed on the high-dependency section of our ward according to strict pre-defined criteria. Oliguria (<0.5 ml/kg per h) was treated with intravenous furosemide (20 mg) and repeated as necessary. Dopamine was started if patients failed to produce an adequate urine output over the subsequent hour despite furosemide. Dopamine was also administered if the systolic blood pressure fell to below 95 mmHg or fell by over 25 mmHg and did not immediately respond to colloid challenge. The necessity for dopamine therapy was taken as a failure of ACE inhibitor treatment, which was discontinued and the inotrope weaned over several hours as tolerated.
Patients with a history of myocardial infarction and with impaired ventricular function were further classified into mild left ventricular dysfunction (3565% ejection fraction by ventriculography) and significant left ventricular dysfunction (<35%). The ventriculogram was carried out within 6 months prior to surgery in all cases and there was no intervening history of myocardial infarction. Patients were monitored with continuous three-lead electrocardiogram, pulse oximetry, hourly urine output and non-invasive Doppler blood pressure recordings. Central venous pressure was also monitored and colloid was administered with the furosemide when this reading was low.
Table 1 shows the patient characteristics in the three groups. There were no significant statistical differences in age, weight (pre-op and at day 2 post-op), sodium concentration at the point of entry into the study, dopamine requirement prior to the point of entry into the study, and incidence of diabetes, hypertension and pre-operative enalapril and diuretic therapy. There was also no difference with regard to left ventricular ejection fraction, both for the mild and the significant dysfunction groups.
|
| Results |
|---|
|
|
|---|
|
|
There were no peri-operative deaths. Morbidity included one occurrence of atrial fibrillation requiring cardioversion and one cerebrovascular accident in the placebo group, one chest infection, one gastrointestinal haemorrhage and two sternal wound infections in the perindopril group and one chest infection in the enalapril group. Post-operative stay on the ward (following one night on intensive care) was 4.2±0.3 days for the placebo group, 4.3±0.3 days for the perindopril group and 4.2±0.2 days for the enalapril group.
Significant left ventricular dysfunction groups (LVEF<35%)
The creatinine level increased after surgery but to a lesser extent than in the moderate dysfunction groups. Recovery of pre-operative levels was not as complete at post-operative day 2 as in the groups with mild left ventricular dysfunction, particularly in the placebo group (
Fig. 1b).
Fig. 2b shows the first dose response in the groups with significant left ventricular dysfunction after placebo, perindopril and enalapril administration. There was no significant hypotension following placebo and perindopril administration, either in systolic or diastolic pressure measurements. Following enalapril administration there was a fall in systolic blood pressure from 114.7±4.6 to 95.0±3.8 mmHg after 4 h (profile comparison: P=0.008) and in diastolic blood pressure from 75.7±4.6 to 58.3±3.1 mmHg after 4 h (profile comparison P=0.001). This significant first dose response resulted in all seven patients in this group requiring cessation of enalapril therapy and administration of dopamine infusion in order to correct hypotension and transient renal impairment. The post-operative furosemide requirement in this group was also greater (Table 1).
With significant ventricular dysfunction, drug intolerance occurred in one of seven patients in the placebo group, in two of seven patients in the perindopril group and in seven of seven patients in the enalapril group. These differences reached statistical significance (P=0.02 between the perindopril and enalapril groups) but because of multiplicities of comparisons these results should be considered with caution.
There were no peri-operative deaths. Morbidity included one occurrence of atrial fibrillation in the placebo group, two occurrences of atrial fibrillation (all requiring cardioversion) and one sternal wound infection in the perindopril group and one chest infection and one resternotomy for haemorrhage in the enalapril group. Post-operative stay on the ward (following one night on intensive care) was 5.3±0.7 days for the placebo group, 5.6±0.5 days for the perindopril group and 5.4±0.9 days for the enalapril group.
| Discussion |
|---|
|
|
|---|
Cardiopulmonary bypass is associated with physiological disturbances that may lead to end-organ hypoperfusion [14]. Non-pulsatile blood flow, inflammatory mediators, increased catecholamine levels, microemboli, and haemolysis all contribute to renal dysfunction [15] [16] [17] [18]. Renal vascular resistance is increased, blood flow is decreased overall and flow distribution is uneven [19] [20]. The incidence of renal dysfunction and dialysis requirement after cardiopulmonary bypass varies widely in different studies [21] [22] [23]. Risk factors associated with post-operative renal dysfunction include advancing years, diabetes, heart failure, re-operations and raised pre-operative serum creatinine level [23].
Myocardial stunning results in a reversible depression of cardiac output, manifested as hypotension, during the first few hours or days following a myocardial ischaemic event [24]. Oliguria and a rise in serum creatinine are common consequences of this phenomenon in the early post-operative setting and are treated with diuretics, inotropes and judicious colloid therapy. In our study serum creatinine levels increased in all groups following surgery, but to a lesser degree in the significant dysfunction groups than in the mild dysfunction groups. The larger dose of dopamine required post-operatively prior to the point of entry into the study (mean: 2.2 vs. 0.5 mg/kg per min) may have contributed to this finding.
In patients with significant myocardial dysfunction, a poor cardiac output together with increased inotropic support may result in metabolic acidosis. Diuretic therapy may help restore homeostasis but results in hyponatraemia as part of a broad electrolyte loss. Hyponatraemia may be exacerbated by ACE inhibitor treatment [25]. These patients stand to benefit maximally from ACE inhibitor treatment in the long term. Unfortunately first-dose hypotension may compound an already low blood pressure and necessitate the resumption of inotropic support. This scenario often leads to discontinuation of ACE inhibitor treatment with probable loss of long-term benefit.
Our study highlighted a significant underusage of ACE inhibitors pre-operatively. Patients' medication is unlikely to be altered by the primary care physician after discharge from hospital. With increased awareness of the advantages of these drugs we expect this situation to improve. However, it would be advantageous for the cardiac patient to be established on such treatment after surgery and then go on to benefit in the long term.
In this study we chose to commence ACE inhibitor treatment on the first post-operative day and to assess haemodynamic tolerance at this critical time. Inotropes are normally discontinued at this point in the patient's progress, while monitoring in a high-dependency bed is such that any problem that may arise is immediately registered and corrective action taken promptly. This provides a window of opportunity that is lost on the second post-operative day when invasive lines are removed. In the past it was customary for us to commence ACE inhibitor treatment on the fourth or fifth post-operative day when myocardial recovery was assumed to have taken place as evidenced by, among other events, a brisk diuresis and a return of the patient's weight to pre-operative levels. However with increasingly short hospital stay this luxury is no longer afforded by most units.
Two ACE inhibitors with known differing blood pressure responses were selected. Enalapril has been widely tested in patients with heart failure [1] [4] and perindopril has been shown to reduce left ventricular hypertrophy [26] and alter peripheral haemodynamics in humans [11] [27] without significant hypotension [28]. After oral administration these two prodrugs rapidly convert to the active metabolites enalaprilat and perindoprilat, with a long pharmacologic half-life [29]. Long-acting ACE inhibitors have been reported to cause more deleterious effects on renal function [30]. However the consequences of these effects are unclear, especially in the long term [28] [31]. The doses prescribed in our study were not in accordance with the manufacturer's suggested starting dose (2 mg for perindopril and 2.5 mg for enalapril) or target upper range (4 mg for perindopril and 20 mg for enalapril). The doses we selected may have rendered an unfair disadvantage on perindopril, but our results suggest that in spite of this regimen this drug was better tolerated. The criterion chosen was equivalent ACE inhibition (2 mg perindopril versus 2.5 mg enalapril) as demonstrated by Squire et al. [10]. In this same study comparing perindopril 2 mg and enalapril 2.5 mg in heart failure patients Squire concluded that perindopril was less likely to cause an acute hypotensive response. Although certain baseline predictors such as initial blood pressure, age, creatinine level, diuretic dose and NYHA class explained some of the variability in response, none of these parameters predicted the risk adequately for clinical purposes. Squire analysed mean blood pressure responses whereas in our study the individual patient's haemodynamic clinical response was analysed.
In our study, intolerance to ACE inhibitors was ultimately determined by an acute adverse event of hypotension and/or oliguria. Although the groups with significant ventricular dysfunction were small (n=7), enalapril administration was uniformly followed by clinical deterioration within 8 h. Perindopril was better tolerated, but because of an uncertain response we would advise vigilance. When administering ACE inhibitors in the early post-operative setting hourly monitoring in a high dependency setting is essential. Daily creatinine levels were not helpful in altering our clinical management as they reflected the clinical course over the previous 24 h and bore no correlation with the acute events.
The results obtained in our study cannot be attributed to a more aggressive pre-operative diuretic regime in any particular group (Table 1). Post-operative weight loss at day 2 was also similar but in the group with the least haemodynamic tolerance (enalapril group with LVEF <35%) furosemide requirement was greater.
We do not routinely measure cardiac output after CABG and therefore cannot comment on the effect of ACE inhibitor treatment on cardiac output in this scenario. Our study demonstrated that some patients developed hypotension following ACE inhibitor treatment and this was associated with oliguria.
The use of dopamine in the treatment of oliguria and acute renal failure is still controversial, and indeed, not without its reported morbidity [32]. There is no convincing evidence that it is effective when administered prophylactically after elective coronary artery bypass grafting [33]. However, in pre-renal acute renal failure dopamine may temporarily improve renal perfusion while the systemic haemodynamics are rectified [32]. Dopamine treatment appeared to be clinically beneficial in our study in that it resulted in an improvement in blood pressure and urine output.
In conclusion our study demonstrates that ACE inhibitors may be administered early after cardiac surgery, to patients with mild left ventricular dysfunction, even in the face of moderate renal impairment, thus paving the way for early patient discharge from hospital on such treatment. If careful monitoring of haemodynamic parameters as well as hourly measurement of urine output is carried out, renal failure can be avoided. However in patients with significant left ventricular dysfunction considerable caution should be exercised. In our study enalapril resulted in a significant drop in blood pressure and a temporary deterioration in renal function which was corrected with dopamine. This first dose hypotensive response was also encountered with perindopril, but in significantly fewer patients. We therefore conclude that perindopril may offer advantages over enalapril in this particular setting.
| References |
|---|
|
|
|---|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
| ANN THORAC SURG | ASIAN CARDIOVASC THORAC ANN | EUR J CARDIOTHORAC SURG |
| J THORAC CARDIOVASC SURG | ICVTS | ALL CTSNet JOURNALS |