From December 2014 to December 2018, 189 patients with overt hypothyroidism who underwent CABG at Fuwai Hospital (Beijing, China) were identified. The exclusion criteria of the hypothyroid group included (1) no thyroid function tests within two weeks before surgery, (2) patients with subclinical hypothyroidism, (3) concurrent with other cardiac surgery, and (4) previous sternotomy (Fig. 1). The exclusion criteria of the euthyroid group were (1) no thyroid function tests within two weeks before surgery, (2) abnormal thyroid function results, (3) history of thyroid or pituitary disease, (4) concurrent with other cardiac surgery, and (5) previous sternotomy. Finally, 6759 euthyroid patients were identified. The hypothyroid patients were then 1:4 matched with euthyroid patients. Medical history and perioperative records were collected. Echocardiography was used to measure left ventricular end-diastolic diameter (LVEDD) and left ventricular ejection fraction (LVEF). Because this was a retrospective analysis of data collected for routine clinical care, individual informed consent was waived by the ethics committee. The study design was approved by the local ethics committee of Fuwai Hospital, Beijing, China (Approval NO. 2019 − 1151; Date of review: June 26, 2019).
Median sternotomy is the preferred approach in all patients. The decision to use cardiopulmonary bypass (CPB) was left to the operating surgeon and was mostly based on the severity and extent of disease of the target vessels. CPB was conducted on moderate hypothermia, and myocardial protection was achieved by using intermittent anterograde hyperkalemic cold blood cardioplegia. The final goal of CABG was to obtain complete myocardial revascularization.
For thyroid function tests, serum TSH, free thyroxine (FT4), total thyroxine (TT4), free triiodothyronine, and total triiodothyronine were measured using electrochemiluminescence immunoassays on a Siemens analyzer. Overt hypothyroidism was defined as a documented history of overt hypothyroidism in the patients’ clinical record, or a TSH level > 4.78 uIU/mL, FT4 < 0.8 ng/dL and TT4 < 4.29 ug/mL without a history of thyroid diseases. Euthyroidism was defined as 0.55 uIU/mL ≤ TSH level ≤ 4.78 uIU/mL, 0.8 ng/dL ≤ FT4 ≤ 1.88 ng/dL and 4.29 ug/mL ≤ TT4 ≤ 12.47 ug/mL at the time of surgery, and no documented history of thyroid disease in the patients’ clinical record.
The primary outcome was postoperative complications, which included impaired wound healing, reintubation, tracheotomy, reoperation, thoracentesis or pericardiocentesis, intra-aortic balloon pump (IABP) implantation, dialysis, postoperative stroke, pacemaker implantation, ventricular arrhythmia, atrial fibrillation, and death. The impaired wound healing was defined as hematoma, wound dehiscence, skin necrosis, lymphatic wound drainage, and local signs of infection. Postoperative stroke was defined as a new neurological deficit with imaging of the central nervous system infarction or hemorrhage. Reoperation was defined as the second operation under general anesthesia during the same hospitalization. Patients were routinely monitored postoperatively for the occurrence of arrhythmia by bedside monitors. When needed, 12-lead electrocardiograms were obtained to determine the exact cardiac rhythm.
The secondary outcomes included postoperative inotropic support and mechanical ventilation time. Postoperative inotropic support included the use of dopamine, adrenaline, noradrenaline, milrinone, and dobutamine. Both the proportion and duration of these drugs were recorded. The total duration of inotropic support was the sum of dopamine duration, adrenaline duration, noradrenaline duration, milrinone duration, and dobutamine duration. When postoperative cardiac output was low and cardiovascular performance was bad, dopamine, dobutamine, or milrinone may be infused initially. When an insufficient response was obtained from these drugs, adrenaline was added or substituted. When hypotension existed in the presence of adequate cardiac output, noradrenaline was used. For patients with overt hypothyroidism, levothyroxine treatment started on the first day after surgery.
Continuous variables are expressed as median (25th, 75th percentile) and were compared in unmatched pairs using the Wilcoxon rank-sum test. Continuous variables in matched pairs were compared using the Wilcoxon signed-rank test. Dichotomous variables are expressed as percentages. The McNemar’s test was applied for dichotomous variables in matched pairs data. Risk factors of postoperative complications were analyzed by condition logistic regression, stratified on the matched sets. Factors for which the univariate analysis gave a p value ≤ 0.1, or of known biologic significance, were included in a logistic multivariate regression model. Odds ratio (OR) and 95% confidence interval (CI) were calculated.
To account for the intergroup clinical imbalance caused by the selection bias inherent in the nonrandomized nature of the study, a continuous propensity score matching (PSM) analysis was performed. A logistic regression model was built to calculate the probability of each patient and to determine the propensity score. Variables in the model included age, gender, weight, myocardial infarction, diseased vessels, previous percutaneous coronary intervention, hyperlipemia, hypertension, diabetes mellitus, stroke, smoking, chronic obstructive pulmonary disease (COPD), preoperative LVEDD, preoperative LVEF, preoperative glomerular filtration rate (GFR), use of CPB and number of distal anastomoses. A greedy matching algorithm was used to perform the 1:4 match on the logit of the propensity score with a caliper width of 0.2. Finally, we matched 187 overt hypothyroid patients with 737 euthyroid patients. The balance of the 2 matched groups was evaluated by standardized mean differences in the matching variables. Usually, a maximum standardized mean difference of 0.1 is considered acceptable.
A p value < 0.05 was considered to be statistically significant. The statistical analysis was performed using SPSS version 22.0 for Windows (SPSS Inc., Chicago, IL, USA).