The Effect of Liver Cirrhosis on Patients Undergoing Cardiac Surgery

Abstract

The aim of this study was to investigate the impact of liver cirrhosis (LC) on postoperative complications and long-term outcomes in patients who underwent cardiac surgery. Three databases, including PubMed, Embase, and the Cochrane Library, were searched on July 24, 2022. A total of 1,535,129 patients were enrolled in the seven included studies for analysis. According to our analysis, LC was a risk factor for postoperative overall complications (OR = 1.48, 95% CI = 1.21 to 1.81, I2 = 90.35%, P = 0.00 < 0.1). For various complications, more patients developed pulmonary (OR = 1.86, 95% CI = 1.21 to 2.87, I2 = 90.79%, P = 0.00 < 0.1), gastrointestinal (OR = 2.03, 95% CI = 1.32 to 3.11, I2 = 0.00%, P = 0.00 < 0.05), renal (OR = 2.20, 95% CI = 1.41 to 3.45, I2 = 91.60%, P = 0.00 < 0.1), neurological (OR = 1.14, 95% CI = 1.03 to 1.26, I2 = 7.35%, P = 0.01 < 0.05), and infectious (OR = 2.02, 95% CI = 1.17 to 3.50, I2 = 92.37%, P = 0.01 < 0.1) complications after surgery in the LC group. As for cardiovascular (OR = 1.07, 95% CI = 0.85 to 1.35, I2 = 75.23%, P = 0.58 > 0.1) complications, there was no statistical significance between the 2 groups. As for long-term outcomes, we found that in-hospital death (OR = 2.53, 95% CI = 1.86 to 3.20, I2 = 44.58%, P = 0.00 < 0.05) and death (OR = 3.31, 95% CI = 1.54 to 5.07, I2 = 93.81%, P = 0.00 < 0.1) in the LC group were higher than the non-LC group. LC was a risk factor for cardiac surgery. Patients with LC who would undergo cardiac surgery should be fully assessed for the risks of cardiac surgery. Similarly, the surgeon should assess the patient’s liver function before surgery. BACKGROUND Liver cirrhosis (LC) is a common disease that kills about 1.03 million people each year [1–2]. Patients with LC often suffered from nutritional damage, immune system dysfunction, coagulation disorders, acute kidney injury, etc. [3–5]. These issues cause surgeons to be hesitant to operate on patients with LC. Surgery on patients with LC remains a challenge for surgeons and anaesthetists. This challenge depends on the type of liver disease and its severity, the surgical procedure, and the type of anaesthesia [6–7]. As for cardiothoracic surgery, especially cardiac surgery requiring cardiopulmonary bypass, LC remains a tricky problem [8–10]. Liver disease remains a major risk factor in the perioperative period of cardiac surgery [11]. Cardiac disease could be a fatal condition. Surgery was an excellent treatment. As for LC patients, the surgeon should assess the patient’s liver function before surgery. The underlying physiological conditions caused by LC make these patients vulnerable to coagulation dysfunction and major organ dysfunction after direct cardiac surgery with extracorporeal circulation [12]. Figuring out the impact and mechanism of LC on cardiac surgery could help surgeons find preventive measures. However, according to our review of previous studies, the effects of LC on the postoperative outcomes of cardiac surgery continued to be controversial. Some studies suggest a poor effect of LC on postoperative outcomes in cardiac surgery [13–17]. Other studies demonstrated that there was no association between LC and postoperative cardiac surgery outcomes [18–19]. Therefore, this pooling-up analysis aimed to investigate the impact of LC on postoperative complications and long-term outcomes in patients who underwent cardiac surgery. METHODS Our meta-analysis was produced in accordance with the Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA) statement [20]. Three databases were searched, including PubMed, Embase, and the Cochrane Library, on July 24, 2022. The key words of the search strategy were LC and cardiac surgery. The search strategy for LC was as follows: “liver cirrhosis” OR “cirrhosis” OR “cirrhotic”, and as for cardiac surgery, we searched “cardiac surgery” OR “cardiac operation” OR “heart surgery” OR “heart operation” OR “thoracic surgery” OR “cardiac surgical procedures” OR “cardiopulmonary bypass” OR “CPB” OR “congenital heart disease”. Then, the two search strategies were combined by “AND”. The search was limited to titles and abstracts, and the language was limited to English. The inclusion criteria for eligible studies were as follows: 1) all patients were diagnosed with cardiac disease and underwent cardiac surgery; 2) both the LC group and the non-LC group were reported; 3) at least 1 of the following complications (cardiovascular, pulmonary, gastrointestinal, renal, neurological, infectious) was reported; and 4) as for long-term outcomes, in-hospital death or death should be reported. The exclusion criteria were as follows: 1) case reports, case series, comments, letters to the editor, conference abstracts, and nonoriginal articles; 2) data was repeated or overlapped; and 3) incomplete information. Two authors searched the databases and identified eligible studies separately. First, duplicate studies were excluded. Then, the two authors scanned the titles and abstracts to find eligible studies. Finally, the full text would be read to identify studies that could be included. Any disagreements were settled by a third author. Patients were divided into the LC group and the non-LC group according to whether they were diagnosed with LC. The cardiac surgery types included coronary artery bypass graft, surgery with cardiopulmonary bypass, cardiac surgery, and aortic valve replacement. The complication was defined as a cardiovascular, pulmonary, gastrointestinal, renal, neurological, or infectious disease that occurred after surgery. Overall complication was the sum of all complications reported in the included studies that were not directly reported. In-hospital death was defined as a patient’s cause of death when they died in the hospital after surgery. Death was defined as a patient’s cause of death after they left the hospital. The information included characteristics of the studies, baseline information on patients, medical history, postoperative complications, and long-term outcomes. The characteristics of the studies were as follows: the first author, published year, published country, study date, sample size of the LC group and the non-LC group, and Newcastle-Ottawa Scale (NOS) score. The baseline information for patients included age, gender, and race. For medical history, hypertension, diabetes, chronic obstructive pulmonary disease (COPD), heart failure, prior myocardial infarction (MI), and malignancy were collected. For complications, cardiovascular, pulmonary, gastrointestinal, renal, neurological, and infectious diseases were collected. As for long-term outcomes, in-hospital death and death were collected. In-hospital death was defined as a death that occurred during hospitalization, regardless of cause. Death was defined as a death that occurred after discharge from the hospital due to cardiac failure or liver cirrhosis. If the cause of death during this period was unknown, it was also considered related. The NOS was used to assess the quality of the included studies [21]. High-quality studies would be scored at nine points. Median quality had scores ranging from 7 to 8 points. Low-quality studies were indicated by a score of less than 7 points. The mean difference (MD) and 95% confidence interval (CI) were calculated for age. The odds ratio (OR) and 95% CI were calculated for gender, medical history, complications, and long-term outcomes. I2 values were used to assess the statistical heterogeneity of the included studies [22–23]. When I2 > 50%, which was considered to be high heterogeneity, the random effects model was used, and P < 0.1 was considered statistically significant. Otherwise, the fixed effects model would be used, and p < 0.05 meant statistically significant. We performed data analysis with Stata V16.0 software. RESULTS A total of 1,158 studies were searched from the three databases (353 studies in PubMed, 778 studies in Embase, and 27 studies in the Cochrane Library). 359 duplicate studies were eliminated. After the remaining 799 studies were viewed for titles and abstracts, 21 studies were left for full-text screening. Then, there were 13 studies left for qualitative synthesis. Finally, seven eligible studies were included in this analysis [13–19]. (Figure 1)