Introduction

Important organs all become vulnerable with aging, therefore, the possibility of cardiac event is increased in elderly patients [1]. Elderly patients with surgical disease have a high prevalence of cardiac risk factors and the stress of abdominal surgery predisposes these patients to major cardiac events [2]. Cardiac events are one of the leading causes of mortality in non-cardiac surgery patients. Among them myocardial infarction is the most common postoperative complication [3].

With the development of more sensitive and myocardial-specific cardiac biomarkers, we can detect small amounts of myocardial injury and necrosis, which improved the definition of myocardial infarction, and highlighted the cardiac biomarker at the same time [4].

Some studies have shown that troponin levels are highly predictive of patient and graft survival following liver transplantation, helping risk stratify patients [5]. Recently, some studies clearified the definition of myocardial injury, extending the range of myocardial infarction to small myocardial injury, and demonstrated that myocardial injury after non-cardiac surgery is common and associated with substantial mortality [6, 7]. To our knowledge, there is no study about myocardial injury after abdominal surgery in elderly patients so far.

We performed this study aiming to identify the incidence, predictors, characteristics and impact on prognosis of MI in elderly patients underwent abdominal surgery.

Materials and methods

After approved by the Beijing Hospital Institutional Review Board(2016BJYYEC-030-01), the retrospective study was carried out at National Center of Gerontology. The medical records of patients ≥ 65 who underwent abdominal surgery longer than 2 h between January 2016 and March 2017 were reviewed.

Cardiac evaluation and treatment were done by cardiac physician preoperatively, cardiac risk factors and cardiac medical history were collected in detail; patients’ nowadays cardiac symptoms and heart function were evaluated by questioning and physical examination; further examinations were performed according to decision of cardiac physician, including echocardiography, stress test even cardiac catheter.

Balanced technique with intravenous and/or inhalation anesthetics combined with opioids and the muscle relaxants was used for anesthetic induction and maintenance. In addition to the standard monitors, the direct arterial blood pressure monitor was used. Intra-operative fluid therapy was fulfilled with crystalloids, colloid, FFP and RBC. Intra-operative vasopressors were used if needed either in bolus or in continuous infusion.

Patients’ preoperative and intra-operative variables were prospectively collected. Postoperatively condition, troponin I, CKMB, myoglobin and BNP level, corresponding situation and resolution were collected until the patients discharged.

MI was defined as prognostic-relevant myocardial injury that occurs within once-administration-period after surgery, manifested by a typical increase in troponin I (≥ 0.04 ng/ml), performed with an electro-chemiluminescent immunoassay (Roche Diagnostics), excluding non-ischemia multi-factorial myocardial injury (e.g., underwent cardiovertion, sepsis or pulmonary embolism). [7] We did not routinely do troponin I measurement for all patients, but only for the patients who had risk factors before, events intra-operative, or symptoms and events after surgery, we hypothetically grouped those patients who had no troponin I measurement to patients without MI.

Myocardial infarction was defined as a typical increase in cardiac biomarker within once-administration-period after surgery and one or more of the following features: symptoms (such as pain of chest, shoulder or arm); ECG ischemia change (ST-segment elevation or depression, T-wave inversion, new left bundle branch block); development of pathologic Q wave; imaging evidence of new loss of visible myocardium or regional wall motion abnormality; Identification of intra-coronary thrombus by angiography or autopsy. [4].

The preoperative, intra-operative and postoperative variables were compared between patients with MI and without MI. The incidence of MI were calculated.

Large dose of vasopressor was defined as using more than two kinds of vasopressor. Infection was defined as fever and WBC ≥ 1.5 times normal upper limit.

All analyses were performed using the Statistical Package for the BMI version 23.0 (SPSS, Inc., Chicago, IL). All reported p values are two sided. The categorical data were analyzed with Pearson’s Chi-square test. Student’s t test or analysis of variance was used to compare the preoperative and intra-operative quantitative variables. The categorical variables were presented as percentages, and the quantitative variables were presented as mean ± standard deviation. The data which were not accord with Gaussian distribution or homoscedasticity were analyzed with nonparametric tests. The cutoff values that were used to divide the quantitative variables were selected by the median, mean, quartiles, or a clinical meaningful value. Variables with a statistically significant relationship to MI in univariate analysis were entered into multivariate logistic regression. Statistical significance was defined as P less than 0.05.

To make the result accurate, we ruled out the patients who had no troponin I measurement within once-administration-period after surgery, comparing the variables between patients with MI and patients with normal troponin I level.

Result

A total of 612 patients (≥ 65)underwent abdominal surgery (longer than 2 h) at Beijing Hospital, National Center of Gerontology between January 2016 and March 2017. After exclusion of 327 without troponin measurement within once-administration-period after surgery, 285 patients were included in the study. The Flow Chart see Fig. 1. The mean patient age was 72 years (range 65–90 years), with troponin I level range from 0.00 to 4.08 ng/ml.

Fig. 1
figure 1

Flow chart. A total of 612 patients were analyzed

Full size image

Among 612 patients studied, 285 patients (46.6%) had troponin I measurement within once-administration-period after surgery, 36 patients (12.6%) developed troponin I elevation, no patient had non-ischemia causes, while only two patients (0.7%) fulfilled definition of myocardial infarction. Most of troponin I, CKMB, Myoglobin and BNP peak appeared within the first 7 days after surgery, the profiles of troponin I and other relative biomarkers after surgery are shown in Fig. 2.

Fig. 2
figure 2

The profiles of troponin I and other relative biomarkers after surgery

Full size image

Results of the univariate analysis of the studied pre- and intra-operative variables between patients with MI and without MI are shown in Tables 1 and 2. Patients with MI had significantly higher incidence of CAD and infection before surgery compared with patients without MI. Many intra-operative variables were associated with MI in univariate analysis, such as laparoscopic/laparotomy surgery, gastrointestinal/hepatobiliary surgery, blood loss and anesthetic maintenance method, etc.

Table 1 Preoperative variables between MI group and No-MI group
Full size table
Table 2 Intra-operative variables between MI group and No-MI group
Full size table

In multivariable analysis, CAD history, non-laparoscopic surgery, blood loss ≥ 800 ml, non-venous maintain, and infection were confirmed as independent risk factors of MI (Table 3). The risk factors had 2.1–5.2 fold increase odds ratio developing MI.

Table 3 multivariable analysis
Full size table

Results of postoperative outcomes comparing between patients with MI and without MI are shown in Table 4. MI were associated with longer hospital stay, more cardiac consultation, higher infection and reoperation rate, the patients with MI also had lower hemoglobin level and higher BUN, K, glucose level.

Table 4 Postoperative variables between MI group and No-MI group
Full size table

Meanwhile, the incidences of mainly outcomes were proportional to the troponin level, the higher of troponin I, more frequency of the outcomes.There was significant difference between troponin ≥ 0.04 ng/ml group and troponin ≥ 0.1 ng/ml group (P < 0.05). See Fig. 3.

Fig. 3
figure 3

Different troponin level results in different outcomes

Full size image

Discussion

In our study, we confirmed that MI was common in elderly patients who underwent abdominal surgery (incidence 12.6%), while myocardial infarction was infrequence (incidence 0.7%). They had similar risk factors and both did harm to the prognosis of elderly patients [8]. Therefore, we should pay more attention not only to myocardial infarction patients but also to MI patients. Failure to monitor troponin measurements after surgery will result in missing most of the prognosis-relevant events.

The new universal definition of myocardial infarction highlights cardiac biomarkers [4, 9]. In our study, we found that 12.6% of our investigated population developed troponin I elevation within once-administration-period after surgery. Cardiac troponin I and T are components of the contractile apparatus of myocardial cells and are expressed almost exclusively in the heart [10]. Other biomarkers such as CKMB, myoglobin and BNP also have their clinical significance, we observed corresponding elevation of them in our study, although not that much speciality to MI like troponin [11,12,13]. Cardiac troponin elevation is still multi-factorial, including ischemia causes and non-ischemia causes, resulting from myocardial injury or other important organ injury [14, 15].

Non-ischemia causes include sepsis, pulmonary embolism, stroke, cardioversion, etc. According to the definition of myocardial injury after non-cardiac surgery, perioperative myocardial injuries due to a documented non-ischemic etiology should be excluded [7]. In our study, we found no non-ischemia cases, that is 100% of myocardial injury cases were ischemia related, which maybe because the limited sample size. Although experienced physicians investigated all elevated troponin I measurements to evaluate if there was evidence of a non-ischemic cause, it is still possible some non-ischemic etiologies were missed and that some events were not due to ischemic myocardial injury.

Previous studies, using troponin T as biomarker, showed that the elevation of an isolated cardiac biomarker or enzyme level during the first 3 days was an independent predictor of 30-day mortality [6, 7]. LeManach et al. [16] conducted a consecutive cohort study of 1136 patients undergoing abdominal aortic surgery in which they excluded non-ischemia patients, they demonstrated that an elevated troponin I after surgery was an independent predictor of in-hospital mortality. While our study evaluated all troponin I elevations within once-administration-period after surgery, and excluded troponin I-elevated cases related to non-ischemia, avoiding missing the delay-onset myocardial injury and observing that majority of myocardial injury cases are ischemia related. We demonstrated that whatever the reasons of troponin I elevation are, troponin I elevation is related to bad prognosis after surgery, deserving more attention and management.

In our study, we only did troponin I measurement for part of the patients who had risk factors before, intra-operative event, or symptoms and events after surgery, and we grouped those patients who had no troponin I measurement to patients without troponin I elevation, which might cause bias when calculated the incidence of MI. To make the result more accurate, we excluded the patients who had no troponin I measurement within once-administration-period after surgery, comparing the variables between patients with troponin I elevation and patients with normal troponin I level.

After abundant variable analysis, we identified CAD history, non-laparoscopic surgery, blood loss ≥ 800 ml, non-venous maintain, and infection as independent risk factors for troponin-I elevation. Patients who have history of CAD are more likely suffering cardiac complications under the stress of surgery, which has been established before [17]. There are more and more guidelines helping us to evaluate and manage cardiac diseases before surgery, making CAD a controllable medical history [18]. Preoperative extended evaluation and management should be made to patients with these risk factors, but intra-operative management seems more important than preoperative factors.

Laparoscopic surgery with minimal trauma puts less stress on patients, whereas laparotomy surgery with massive tissue injury gives large burden on patients’ important organs. The laparotomy surgery will five times increase MI incidence, which is similar with previous study [19, 20].

In the present study, we confirmed that blood loss ≥ 800 ml would triple the possibility of MI, pointing out that the more blood loss occurred, the more myocardial injury would happen, suggesting that blood loss should be kept to a minimum during the surgery [21].

Venous maintenance is commonly used during anesthesia, to avoid pollution, nowadays venous anesthetics are applied more frequency than inhalation anesthetics [22]. Although previous study has shown that sevoflurane has myocardial protect effect, [23] in our study, we observed that solely inhalation maintenance doubled MI incidence. The reason maybe combination remifentanil and propofol, effectively depressed the stress from surgical stimuli, decreased HR and cardiac oxygen consumption, therefore, prevented MI onset [24]. Further studies about balancing the advantage and disadvantage of inhalation anesthetics and venous anesthetics are needed.

Infection is the leading cause of mortality after surgery,[3] systemic inflammation reaction will do harm to important organs especially heart and kidney, added with increased HR, heart burdened a lot, almost five times increased MI incidence, controlling infection is of great importance [25].

In our study, we also observed that most of the troponin I elevation happened in the first 7 days after surgery, with other related biomarker accordingly elevated, but only small amount of patients (0.7%) suffering troponin I elevation experienced an ischemic symptom. Therefore, most of the troponin I elevation cases probably would have been missed without systematic biomarker monitoring after surgery. Moreover, the higher the troponin I level was, more frequently happened the bad outcomes. Because MI is common, hidden and leads to bad prognosis, clinical trials to establish strategies to prevent and treat this important complication were urgently needed. We recommend routinely monitoring perioperative troponin I measurements in elderly patients undergoing abdominal surgery, especially on the first 7 days after surgery for the high-risk population.

Our study has some limitations. First, this is a retrospective study, the question remains whether the risk factors identified in this study were surrogate markers of myocardial injury. In addition, the possibility of some other unmeasured confounding factors should be considered. Second, the results of our study were based on data from one center, while there are a variety of perioperative practices during surgery among different centers. Therefore, our results need to be confirmed by other centers. Further prospective studies designed to prevent and treat MI are warranted.

In conclusion, MI is common in elderly patients undergoing abdominal surgery, which has negative impact on prognosis. Perioperative troponin I evaluation in elderly patients, especially within the first 7 days after surgery, may aid in the risk stratification and prognostic assessment. Decreasing intra-operative blood loss, laparoscopic surgery, venousmaintenance anesthesia, control infection, and focus on patients with CAD may help preventing post-operative myocardial injury. Preventative medicine and effective management of MI need further investigation.