|
|
||||||||
Castle Hill Hospital, Kingston-Upon-Hull, East Yorkshire, United Kingdom
Received 1 September 2007; received in revised form 10 January 2008; accepted 11 January 2008.
* Corresponding author. Address: Department of Cardiothoracic Surgery, Castle Hill Hospital, Kingston-Upon-Hull, East Yorkshire HU16 5JQ, United Kingdom. Tel.: +44 1482 623256; fax: +44 1482 623257. (Email: dngaage{at}yahoo.com).
| Abstract |
|---|
|
|
|---|
80 years. Neurological complications, classified as reversible or permanent, were investigated by head CT scan in patients who did not recover soon after an event. Results: There were more females (47% vs 26%, p
< 0.0001) among octogenarians (n
= 383, median age 82 years) than among younger patients (n
= 6408, median age 66 years). Controlled heart failure, NYHA class III/IV and chronic obstructive pulmonary disease were more prevalent in octogenarians while preoperative myocardial infarction was predominant in younger patients. Octogenarians were at higher operative risk (median EuroScore 6 vs 2, p
< 0.0001). Operative procedures differed between octogenarians and younger patients (p
< 0.0001); respective frequencies were 45% vs 77% for CABG, 26% vs 10% for AVR, and 23% vs 6% for AVR + CABG. Mortality was higher for octogenarians (8.9% vs 2.1, p
< 0.0001). Early neurological complications observed in 3.9% of the entire study population were mostly reversible (3.2%). Age
80 years (odds ratio [OR] 2.82, 95% confidence interval [CI] 1.89–4.21, p
< 0.0001), prior cerebrovascular disease (OR 2.23, 95% CI 1.56–3.18, p
< 0.0001), AVR + CABG (OR 2.92, 95% CI 1.60–5.33, p
< 0.0001) and MVR + CABG (OR 4.77, 95% CI 2.10–10.85, p
< 0.0001) were predictive of neurological complications. More octogenarians experienced neurological events (p
< 0.0001): overall 12.8% vs 3.4%, reversible 11.5% vs 2.8%, permanent 1.3% vs 0.6%. Among octogenarians, neurological complication was associated with elevated operative mortality (18% vs 8% for those without neurological complication, p
= 0.03), and prolonged ventilation, intensive care stay and hospitalisation. Predictors of neurological complications in octogenarians were blood and/or blood product transfusion (OR 3.60, 95% CI 1.56–8.32, p
= 0.003) and NYHA class III/IV (OR 7.6, 95% CI 1.47–39.70, p
= 0.02). Conclusion: Octogenarians undergoing on-pump CABG and/or valve repair/replacement are at higher risk of neurological dysfunction, from which the majority recover fully. The adverse implications for operative mortality and morbidity, however, are profound. Blood product transfusion which has a powerful correlation with neurological complication should be reduced by rigorous haemostasis with parsimonious use of sealants when appropriate.
Key Words: Octogenarian Cardiac surgery Mortality Neurological complications
| 1. Introduction |
|---|
|
|
|---|
The objectives of this study were therefore to (1) report the incidence of early neurological complications after on-pump coronary artery bypass grafting (CABG) and/or valve replacement or repair, (2) identify preoperative and procedural risk factors for these complications, and (3) assess the influence of these factors on postoperative outcome in octogenarians. We have also provided matching data for a contemporaneous population of patients younger than 80 years old who underwent similar operations at same institution to put our report in perspective.
| 2. Materials and methods |
|---|
|
|
|---|
80 years of age. The local medical and ethics committee approved the use of patient clinical data for this study in November 2006. The cardiothoracic surgery database was interrogated to obtain prospectively collected preoperative, operative and postoperative data. Head CT scan reports of all patients investigated for postoperative neurological dysfunction were retrieved. Where necessary, charts were further reviewed to obtain additional data. Early postoperative neurological events were broadly classified as reversible or irreversible, based on the findings of clinical evaluation at the time of the event and at hospital discharge. The reversible neurological events observed in the study patients were confusion/agitation, seizures and transient ischaemic attack (TIA) or reversible ischaemic neurological deficit (RIND). Residual deficit at hospital discharge or coma were categorised as permanent or irreversible neurological event.
2.2 Data analysis
Statistical analysis was performed using the Statistical Package for the Social Sciences (SPSS) version 14.0 for Windows, (SPSS Inc. 2005, Chicago, IL). Categorical variables are reported as percentages and compared between octogenarians and younger patients with the chi-square test, while continuous variables are reported as median with 25th and 75th percentiles as interquartile range (IQR), and compared between groups using the Mann–Whitney U test. Independent predictors of early adverse neurological outcome and mortality in the entire study population were identified using a stepwise multi-factorial logistic regression model constructed with all the preoperative and operative variables in Table 1
. Further analysis was performed for octogenarians to determine the effect of postoperative neurological event on operative mortality and length of hospital stay. Fig. 1
illustrates the tiers and groups of patient included in the analyses. A two-sided p
< 0.05 was considered significant.
|
|
| 3. Results |
|---|
|
|
|---|
3.2 Operative outcome
Compared to younger patients, octogenarians had higher predicted operative mortality (median EuroScore of 6, IQR 8–9 vs 2, IQR 3–5, p
< 0.0001) and more often their operations were performed non-electively (39% vs 22%, p
< 0.0001). Whereas coronary artery disease was the single most common indication for surgery in the younger age group, octogenarians had surgery mainly for valvular heart disease with or without coronary artery disease. As a result there are striking differences in the types of operation performed between the two groups (p
< 0.0001). The corresponding proportion of operative procedures for octogenarians compared to younger patients were 45% vs 77% for isolated CABG, 26% vs 10% for isolated aortic valve replacement (AVR) and, 23% vs 6% for AVR + CABG. The overall operative mortality was significantly higher for octogenarians (8.9% vs 2.1%, p
< 0.0001). For the different types of operations, octogenarians had greater operative death rates as illustrated in Table 1.
The predictors of operative mortality were female gender (odds ratio [OR] 1.93, 95% confidence interval [CI] 1.15–3.25, p = 0.01), emergency operation (OR 4.10, 95% CI 1.53–10.98 p = 0.005), controlled heart failure (OR 1.82, 95% CI 1.05–3.16, p = 0.03), history of cerebrovascular disease (OR 2.22, 95% CI 1.27–3.93, p = 0.006), left ventricular ejection fraction < 30% (OR 3.97, 95% CI 2.0–7.91, p < 0.0001), duration of cardiopulmonary bypass (OR 1.009, 95% CI 1.004–1.014, p < 0.0001), reopening for bleeding (OR 6.34, 95% CI 3.38–11.51, p < 0.0001), postoperative low cardiac output state (OR 6.69, 95% CI 3.24–13.83, p < 0.0001), readmission to intensive care unit (OR 8.84, 95% CI 5.0–15.63, p < 0.0001) and postoperative neurological complication (OR 3.34, 95% CI 1.81–6.16, p < 0.0001).
3.2.1 Neurological complications
The incidence of early postoperative neurological complication in the entire study population was 3.9%; reversible events accounted for 3.2% (confusion/agitation/seizures 2.2%, TIA/RIND 1%) and permanent deficit for 0.7%. Octogenarians experienced substantially higher rates of both reversible (11.5% vs 2.8%, p
< 0.0001) and permanent (1.3% vs 0.6%, p
< 0.0001) neurological events than younger patients (Fig. 2
).
|
80 years (OR 2.82, 95% CI 1.89–4.21, p
< 0.0001), AVR + CABG (OR 2.92, 95% CI 1.60–5.33, p
< 0.0001) and mitral valve replacement/repair (MVR) + CABG (OR 4.77, 95% CI 2.10–10.85, p
< 0.0001).
3.2.2 Other postoperative complications
Postoperative complications occurred in more octogenarians than younger patients (68% vs 50%, p
< 0.0001). A breakdown of the complications is displayed In Table 1.
The respective rates for readmission to intensive care unit (9% vs 3%, p < 0.0001) and discharge to an intermediate care facility for continued management and/or supervised recovery (21% vs 7%, p < 0.0001) were markedly greater for octogenarians than younger patients. The median length of postoperative hospital stay (7 days, IQR 9–13 vs 6 days, IQR 6–8, p < 0.0001) was also longer for octogenarians.
3.3 Sub analysis of octogenarians
We further analysed the operative outcome in octogenarians. Four patients were
90 years of age. The operations performed in majority of octogenarians were CABG and AVR either as isolated or concomitant procedures, or combined with MVR. One patient underwent CABG and AVR combined with MVR and tricuspid valve repair.
3.3.1 Operative mortality and morbidity
The operative mortality rates were 8.6% for CABG, 5.1% for AVR, 7.9% for combined AVR and CABG, 15% for MVR, and 57% for combined MVR and CABG. Cardiac causes including failure to wean from cardiopulmonary bypass machine, ventricular arrhythmia and myocardial infarction accounted for majority of operative deaths.
Cardiac complications such as arrhythmia (52%) and low cardiac output state (45%) were the predominant causes of postoperative morbidity. About half of the patients (n = 214, 56%) received blood and/or blood product transfusion. Wound infection (graft donor site and sternal) was observed in 37 patients (10%) and debridement was necessary for 10 (2.6%). Two patients had mediastinitis.
3.3.2 Neurological complications
Neurological complications were observed in 49 patients (13%) and in the vast majority neurological impairment was reversible; only 5 patients (1.3%) had residual neurological deficit at the time of hospital discharge. Confusion/agitation/seizure was observed in 31 patients (8.1%) and TIA/RIND in 13 (3.4%). The predictors of early postoperative neurological complications in octogenarians were transfusion of blood and/or blood products (OR 3.60, 95% CI 1.56–8.32, p
= 0.003) and NYHA class III/IV (OR 7.6, 95% CI 1.47–39.70, p
= 0.02).
3.4 Influence of postoperative neurological events on operative outcome in octogenarians
The baseline characteristics were comparable between octogenarians with neurological event and those without. Operative mortality and morbidity were substantially higher among octogenarians who experienced an acute postoperative neurological event (Table 2
). The median length of stay in intensive care unit (46 h, IQR 22–115 vs 24 h, IQR 22–42, p 0.0001) and duration of postoperative hospitalisation (12 days, IQR 9–18 vs 7days IQR 9–13, p
< 0.0001) were extensively longer for octogenarians with neurological complications.
|
Head CT scan was performed in 17 patients who did not recover full neurological function soon after the event. Fourteen of these had neurological deficit but CT scan delineated ischaemic cerebral infarct in two patients with TIA/RIND (n = 13) and four with permanent deficit (n = 5). The lesions were located in the left cerebral hemisphere in all but one patient. None of the patients with confusion or agitation or seizure had a brain lesion.
| 4. Discussion |
|---|
|
|
|---|
Older age (
80 years) by itself is a powerful predictor of this outcome. Valve operations, more frequently performed in octogenarians also increases the risk of acute postoperative neurological dysfunction. Weintraub et al. [2] examined the influence of age on the outcome of CABG in 13,625 patients, of whom 146 were octogenarians and reported a higher incidence of neurological injury in older patients and a powerful correlation between neurological injury and advanced age. Other series [1,5,9] have also reported a strong association of age and postoperative neurological dysfunction after CABG. The neurocognitive vulnerability of octogenarians is multi-factorial. The age-related morphological and physiological changes characterised by cerebral atrophy and diminished cerebrovascular reserve capacity respectively, combined with a high prevalence of sub-clinical degenerative brain disease and cerebrovascular atherosclerosis contribute to the high propensity to develop acute postoperative neurological dysfunction. The strategies to reduce this complication include attenuating systemic inflammatory response to cardiopulmonary bypass, minimising cerebral embolisation and avoiding cerebral hypoperfusion without cerebral over-perfusion which can be counterproductive [10]. Off-pump, especially with no touch aorta technique [11–14], has been credited with reduced rates of neurological complications following CABG. Similarly, modifying the cardiopulmonary bypass in conjunction with epiaortic scanning [9,15] and processing mediastinal shed blood with cell saver [16] have been associated with improved outcome.
Amongst octogenarians, postoperative blood and/or blood product transfusion increases the risk of acute postoperative neurological dysfunction. Granted the higher risk of postoperative bleeding in octogenarians as evidenced by the prominent use of the anti-fibrinolytic agent, aprotinin, higher blood and/or blood product transfusion rate, and greater incidence of mediastinal re-exploration for control of bleeding in our study, and corroborated by other data [17], particular attention should be directed to meticulous haemostasis, including a parsimonious use of sealants when necessary. It can be argued, quite creditably, that transfusion of blood and blood products does not have a direct causative relationship with postoperative neurological dysfunction. In octogenarians, who have a diminished cerebral reserve, postoperative hypovolaemia for which transfusions are required, may lead to cerebral hypoperfusion and neurological dysfunction. This further emphasises the importance of reducing bleeding in octogenarians. Additional precautions should be taken for octogenarians with a history of cerebrovascular disease and those undergoing combined valve and CABG procedures who are at higher risk, as this study and others [18–20] have shown.
Acute postoperative neurological events considerably affect the outcome of cardiac surgery in octogenarians. Even though acute cerebral event was directly implicated as the primary cause in 7% of operative deaths, it contributed to 25% of the operative deaths. There were more operative deaths among octogenarians who sustained a postoperative neurological event. For the entire study population including patients of all ages, there was a direct relationship between acute postoperative neurological events and operative mortality. It also increases the tendency to develop respiratory complications, and prolong hospital stay by slowing patient recovery and ambulation. These findings are consistent with other reports [5,8,21–23]. However, unlike other studies the rates of discharge to intermediate care for convalescence was not different between octogenarians with neurological complications and those without because the majority of patients with neurological impairment recovered fully or well enough to be discharged home.
4.1 Study limitations
The retrospective nature of this study introduces data bias, as we were limited to only the data entered in the database and patient clinical notes. We did not have access to some data that other series have reported to affect the incidence of postoperative neurological impairment. For example intraoperative haemodynamic parameters (9) and the atherosclerotic burden/manipulation of the ascending aorta which can predispose to postoperative neurological events were not investigated in this study, hence the findings of this study does not represent a comprehensive report of the risk of neurological complications after cardiac surgery. Nevertheless, it highlights the independent effect of preoperative and other procedural risk factors.
The impact of non-surgical factors like medications that can precipitate neurological dysfunction, especially in elderly patients, and consequently increase the rates of neurological complications were not determined. However, this would not affect the rate of permanent neurological deficit that is more devastating.
Age has a prominent effect on outcome of cardiac surgery [24], hence its use in all conventional and internationally accepted risk scoring systems in cardiac surgery, like the EuroScore and Parsonnet score. Patient groups defined by different age profiles differ fundamentally and it is very difficult to establish equivalence by statistical permutations, hence we did not match octogenarians with younger patients.
| 5. Conclusion |
|---|
|
|
|---|
80 years, history of cerebrovascular disease and valve combined with CABG operations places patients at a high risk of early postoperative neurological events. Although most of these events are reversible, they have a profound impact on operative mortality and morbidity in octogenarians. Among octogenarians, blood and/or blood product transfusion has a correlation with acute postoperative neurological dysfunction. Therefore, additional to intraoperative measures that ensure adequate cerebral perfusion, minimise embolisation and ameliorate systemic inflammatory response, haemostasis should be meticulous and combined with parsimonious use of sealants when appropriate. | Appendix A |
|---|
|
|
|---|
Dr H. Jakob (Essen, Germany): You presented the world's best results on neurological outcome in octogenarians, so you have to be complimented for that. On the other hand, this raises a couple of questions.
You have just seen 5 out of 383 patients having a complete neurological deficit. This means stroke. The retrospective nature of this study might be one concern.
So my first question is, why didnt you ask your colleagues from neurology to perform the study probably reducing the bias a little bit since, as we all know, we are more self-forgiving than the other specialists looking for our results.
In addition, do you have any explanation for your superb results besides statistical comparison? I mean, what are you doing surgically which is different from the rest of us?
Third, how do you deal with heavily calcified aortas or even the porcelain aorta?
And this leads to my final question. What are your patient selection criteria? Whom do you deny operation? Is it based on EuroScore or STS score or on clinical judgment?
Dr Ngaage: This study, like you have said, is a retrospective study, based on data that were prospectively collected. The diagnosis of neurological complication is a clinical diagnosis further investigated by CT scan in 17 patients who did not recover early enough, and for who we had concerns. And in those 17, all the patients who had reversible complications like confusion, did not have any cerebral lesion. Four of the 5 patients who had permanent deficit and, 2 of the patients who had TIAs had ischaemic lesions.
We did not involve neurologists in all cases. For example, reversible deficit like confusion which often occurs late in the evening and at night was diagnosed by the on-site clinician and this is usually documented.
So I do take the point that we did not involve them, but we did have enough information from the clinical evaluation and the CT scan to be able to make this diagnosis.
Our results are good and compare well with other reports in the literature, some of which report a stroke rate of between 1.6 and 3%. Ours is about 1.3%.
About our strategies. We have just started using the mini-bypass system to reduce the systemic inflammation, and while I didnt show it here, we used aprotinin in most of these patients. And maybe that decreases the need for blood transfusion, but I can only speculate. I cannot be certain about that.
About the porcelain aorta, there were two patients not in this series that had porcelain aorta, and one was treated with off-pump, and I think the other was treated with deep hypothermia and circulatory arrest, but those were younger patients and were not included in this study.
We use the EuroScore to risk stratify patients, but we do not have selection criteria. If a surgeon feels that a patient is high risk, the surgeon might decline the patient. But the cardiologists will try the next surgeon, and sometimes patients who have been declined come back to have surgery under a different surgeon because there are no strict criteria for selection.
Dr J. Vaage (Oslo, Norway): Ill try to make this very short. You say that blood transfusions are a risk factor. I dont think you have a scientific background to draw that conclusion. You just show that there is a connection in the same patients that have more blood transfusions.
And another conclusion is that some operations are more difficult and have more complications, and they have also more blood transfusions.
Dr Ngaage: The reason that conclusion was included is because in octogenarians we found blood transfusions and the use of blood product as an independent predictor of neurological complication.
Dr Vaage: I dont think you can draw that conclusion because you just found that it occurred in the same patients.
Dr Ngaage: Im sorry. Im not very sure what your point is.
Dr Vaage: You just found that there are more blood transfusions in the patients who had neurological complications, but you cannot say that this is a cause of neurological complications.
Dr Veit: That's real. But I think this is getting a little bit outside of the topic. For the interest of time, we have to proceed to the next presenter.
| Footnotes |
|---|
Presented at the 21st Annual Meeting of the European Association for Cardio-thoracic Surgery, Geneva, Switzerland, September 16–19, 2007. | References |
|---|
|
|
|---|
This article has been cited by other articles:
![]() |
N. Piazza, Y. Onuma, P. de Jaegere, and P. W. Serruys Guidelines for reporting mortality and morbidity after cardiac valve interventions--need for a reappraisal? Ann. Thorac. Surg., February 1, 2009; 87(2): 357 - 358. [Full Text] [PDF] |
||||
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| 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 |