Impact of diabetes on outcomes of cardiac surgery in a multiethnic Southeast Asian population

Abstract

Although diabetes is rapidly increasing in Asia and has been shown to be associated with worse cardiac surgery outcomes, no research has been done to study the impact of diabetes on cardiac surgery outcomes in a Southeast Asian cohort. Hence, this study aims to delineate the predictors and impact of diabetes after cardiac surgery in a multi-ethnic Southeast Asian cohort. We analysed data from 2831 adult patients undergoing elective cardiac surgery, from 2008 to 2010 in Singapore. Diabetes was found to significantly increase the odds of intensive care unit readmission by 1.70 (95% confidence interval 1.171–2.480, p = 0.005), postoperative infection by 1.73 (95% confidence interval 1.003–2.976, p = 0.049), acute kidney injury by 1.36 (95% confidence interval 1.137–1.626, p = 0.001), postoperative hyperglycaemia by 6.00 (95% confidence interval 4.893–7.348, p < 0.001), and new need for dialysis by 1.71 (95% 1.086–5.360, p = 0.021). In conclusion, diabetes is associated with increased risk for renal dysfunction, hyperglycaemia, and infection after cardiac surgery, similar to the relative risks of diabetes patients observed in Western populations.

Keywords

Diabetes outcomes Southeast Asian

Introduction

Diabetes mellitus is a major health problem worldwide today and is estimated to affect 8.4% of the world’s total population.¹ The majority of diabetes patients in the world live in Southeast Asia and the prevalence of the disease is increasing more rapidly in Asia than elsewhere.^1,2 The prevalence of diabetes in patients undergoing cardiac surgery has been reported to be about 20% in Western cohorts³ and is expected to be higher in Asian cohorts, given the higher prevalence of diabetes in Asian populations.¹ Furthermore, research has shown that diabetes is an important risk factor for coronary artery disease and is associated with poorer cardiac surgery outcomes in Western populations due to the poor quality of target vessels and comorbidities of diabetes.^3,4

Yet, although diabetes has been shown to be associated with worse cardiac surgery outcomes and has a higher prevalence and ethnic association in Southeast Asia, no research has been done to study the impact of diabetes on cardiac surgery outcomes in the local Southeast Asian cohort. Hence, study aims to identify the predictors and impact of diabetes on perioperative cardiac surgery outcomes.

Methods

Study population

With SingHealth Centralised Institutional Review Board (CIRB) (IRB number 2008/137/D) approval and written informed consent from the patients before enrolment, a total of 3008 adult patients undergoing elective cardiac surgery were retrospectively included in this study. Of this, 117 patients were excluded from the study as they belonged to various ethnic groups apart from the three main ethnic groups in Singapore that were being studied: Chinese, Malay, and Indian. Of the remaining 2831 patients analysed, 46.7% (n = 1321) had diabetes. Patients with a known history of diabetes, preoperative random plasma glucose greater than 200 mg/dL or taking antidiabetic medications, or haemoglobin A1c (HbA1c) greater than 6.5% were analysed as diabetes patients.

Preoperative patient characteristics are shown in Table 1 with comparisons between patients with and without diabetes undergoing elective cardiac surgery. The average age was 59.4 ± 10.5 years and most patients were male (80.1%) and hypertensive (75.3%).

Table 1.

Preoperative patient characteristics of patients with and without diabetes undergoing cardiac surgery.

Perioperative variables	All (n = 2831)	Patients with diabetes (n = 1321)	Patients without diabetes (n = 1510)	p-value
Preoperative patient characteristics
Female %(n)	19.9 (562)	21.3 (282)	18.5 (280)	0.062
Weight, kg, mean ± SD	66.6 ± 12.7	67.6 ± 13.1	65.8 ± 12.2	<0.001
Height, cm, mean ± SD	163.3 ± 8.2	162.8 ± 8.1	163.7 ± 8.2	0.002
BMI, kg/m², mean ± SD	24.9 ± 4.1	25.4 ± 4.2	24.5 ± 4.0	<0.001
Age, years, mean ± SD	59.4 ± 10.5	60.6 ± 9.0	58.3 ± 11.5	<0.001
Chinese %(n)	71.6 (2026)	65.6 (867)	76.8 (1159)	<0.001
Malay %(n)	16.2 (460)	17.0 (225)	15.6 (235)
Indian %(n)	12.2 (345)	17.3 (229)	7.7 (116)
Preoperative glucose, mmol/L	7.2 ± 3.2	8.7 ± 3.8	5.9 ± 1.8	<0.001
HbAlc, %	11.3 ± 2.7	7.6 ± 2.1	5.4 ± 1.7	<0.001
Highest intraoperative glucose, mmol/L	8.4 ± 5.2	8.8 ± 5.1	8.1 ± 5.3	<0.001
Lowest intraoperative glucose, mmol/L	6.5 ± 3.4	6.8 ± 3.3	6.2 ± 3.4	<0.001
Hypertension %(n)	75.3 (2132)	88.5 (1169)	63.8 (963)	<0.001
eGFR calculator	78.1 ± 25.1	74.4 ± 27.1	81.3 ± 22.7	<0.001
Renal impairment %(n)	12.5 (353)	16.9 (223)	8.6 (130)	<0.001
Serum creatinine >200 µmol/L %(n)	4.5 (128)	7.0 (92)	2.4 (36)	<0.001
LVEF <30 %(n)	8.4 (239)	11.0 (145)	6.2 (94)	<0.001
History of previous cardiac surgery %(n)	2.6 (75)	2.0 (26)	3.2 (49)	0.035
EuroSCORE (logistic), mean ± SD	3.9 ± 5.4	4.1 ± 5.7	3.7 ± 5.1	0.018
CABG surgery %(n)	82.3 (2459)	93.2 (1305)	72.7 (1154)	<0.001
Valvular surgery %(n)	23.7 (712)	11.9 (167)	34.1 (545)	<0.001
Preoperative patient medications
Beta blockers %(n)	74.6 (2240)	82.0 (1151)	68.1 (1089)	<0.001
ACE inhibitors %(n)	48.6 (1459)	56.0 (786)	42.1 (673)	<0.001
Calcium channel blockers %(n)	19.5 (588)	22.4 (315)	17.0 (273)	0.001
Statin %(n)	80.4 (2413)	90.0 (1263)	72.1 (1150)	<0.001
Insulin %(n)	9.1 (274)	19.3 (271)	0.2 (3)	<0.001
OHGA %(n)	35.7 (1069)	74.1 (1040)	1.8 (29)	<0.001
IV diuretics %(n)	18.5 (556)	20.2 (284)	17.0 (272)	0.024
Inhaled steroids %(n)	1.4 (42)	1.3 (18)	1.5 (24)	0.612
Oral steroids %(n)	1.9 (56)	1.9 (26)	1.9 (30)	0.963
Chemotherapy %(n)	0.3 (9)	0.3 (4)	0.3 (5)	0.890

SD: standard deviation; CABG: coronary artery bypass grafting; ACE: angiotensin-converting enzyme; BMI: body mass index; LVEF: left ventricular ejection fraction; EuroSCORE: European System for Cardiac Operative Risk Evaluation; OHGA: oral hypoglycaemic agent; eGFR: estimated glomerular filtration rate; HbAlc: haemoglobin A1c.

Perioperative anaesthesia and surgical and perfusion management

Perioperative clinical practices and surgical management followed strict international standards. Anaesthesia was induced with intravenous induction agents (Propofol or Etomidate) and maintained with a balanced anaesthesia regime using low-dose fentanyl (10–20 μg kg⁻¹) in addition to volatile agents (primarily Sevoflurane). Conventional cardiopulmonary bypass (CPB) circuits with heat exchangers, roller pumps, membrane oxygenators, arterial blood filters, venous reservoirs, and cardiotomy suction were utilised. Typically, 1300–1400 mL of primer was used in the CPB circuits. Perfusion targets include a hematocrit of >22%, mild-to-moderate hypothermia (32°C–35°C), nonpulsatile flow rate of 2.2–2.4 L min⁻¹ m⁻², mean arterial pressure of 50–70 mmHg, glucose levels of below 10 mmol/L, and activated clotting times of more than 400 s. Cold blood cardioplegia was used to achieve myocardial protection. Aprotinin was not utilised in any patients.

All patients were kept to a tight baseline blood glucose control within a range of 4–10 mmol/L, in accordance with hospital protocol and the ‘Society of Thoracic Surgeons Practice Guideline series: Blood Glucose Management During Adult Cardiac Surgery’.⁵ In line with hospital protocol, glycaemic targets were strictly adhered to via insulin infusion and monitoring every 2–4 h or more frequently if required, up to the first 48 h.

Data collected

Data extracted from the patients were chosen a priori and include patient characteristics, risk factors, surgical data, and postoperative outcomes up to 30 days post-surgery including intensive care unit (ICU) mortality, readmission to ICU, length of stay in ICU, length of stay in high dependency unit (HDU), new onset cardiac arrhythmia, cardiac failure, cardiac tamponade, postoperative infections, and length of hospitalisation.

Perioperative outcomes were defined and categorised according to the ICD-10 coding (the 10th revision of the International Statistical Classification of Diseases and Related Health Problems) and noted from patient discharge summaries.

Postoperative hyperglycaemia was defined as a blood glucose level of more than 10 mmol/L (approximately 180 mg/dL) within the first 48 perioperative hours.⁶ The remaining patients who were included were considered normoglycaemic for the purposes of this study.

Acute kidney injury (AKI) was determined based on the Acute Kidney Injury Network (AKIN) criteria and defined as an absolute increase in serum creatinine of ⩾26.4 mmol/L, and/or a ⩾50% increase in the serum creatinine to the peak postoperative from the preoperative serum creatinine. The estimated glomerular filtration rate (eGFR) was calculated via the Chronic Kidney Disease Epidemiology Collaboration (CKD-EPI) equation, during the preoperative anaesthesia review, which has been previously validated in the local population.⁷

Subsequently, we followed the reference cut-off for the eGFR in defining renal insufficiency:^8,9 Stage 1 (representing normal kidney function with eGFR of ⩾90 mL/min/1.73 m²); Stage 2 (mild renal insufficiency with eGFR of ⩾60–89 mL/min/1.73 m²); Stage 3 (moderate renal insufficiency with eGFR of ⩾ 30–59 mL/min/1.73 m²); Stage 4 (severe renal insufficiency with eGFR of ⩾15–29 mL/min/1.73 m²); and Stage 5 (renal failure or end-stage renal disease [ESRD]) with eGFR of <15 mL/min/1.73 m².

In accordance with the new American College of Cardiology/American Heart Association high blood pressure (BP) guidelines in the United States,¹⁰ hypertension was defined as having at least stage 1 hypertension with systolic pressure between >130 mmHg and/or diastolic pressure >80 mmHg.

In accordance with the ‘Definitions of Terms of The Society of Thoracic Surgeons National Cardiac Surgery Database’,¹¹ an infection was considered present if (1) a wound was opened with excision of tissue (I&D), (2) there was a positive culture, or (3) the patient was treated with antibiotics. This included infections that were sternum-superficial (skin, subcutaneous tissue), sternum-deep (muscle, mediastinum, bone), leg (incision site), Intra-Aortic Balloon Pump (IABP) site, and septicaemia (requires positive blood cultures).

Statistical analysis

Continuous variables are expressed as mean ± standard deviation with categorical data as proportions (sample size). For continuous variables, comparisons between patients with and without diabetes were carried out using student’s t-test. Categorical variables were compared using χ² analysis.

The association of the diabetes with adverse postoperative outcomes was then evaluated using logistic regression analysis. Significant patient characteristics, including diabetes and surgical risk factors, were added to the model, and non-significant covariates were removed from the multivariate analysis in a stepwise manner. All statistical analyses were performed using IBM SPSS v.22.0 (SPSS IBM, New York, USA); significance was judged at p = 0.05.

Results

Patient characteristics

Patients with diabetes undergoing cardiac surgery were significantly more obese [body mass index (BMI) 25.4 ± 4.2 vs 24.5 ± 4.0; p < 0.001], older (60.6 ± 9.0 years vs 58.3 ± 11.5 years; p < 0.001), hypertensive (88.5% vs 63.8%; p < 0.001), had renal impairment (16.9% vs 8.6%; p < 0.001), severely reduced left ventricular ejection fraction (11.0% vs 6.2%; p < 0.001) and a higher EuroSCORE (4.1 ± 5.7 vs 3.7 ± 5.1; p = 0.018) (Table 1). Expectedly, patients with diabetes also had higher preoperative glucose (8.7 vs 5.9 mmol/L, p < 0.001), HbA1c (7.6 vs 5.4%, p < 0.001), and peak intraoperative glucose level (8.8 vs 8.1 mmol/L, p < 0.001) (Table 1).

The overall prevalence of diabetes in our cohort was 46.7% (n = 1312 out of 2831) with prevalence varying among the different ethnic groups. Race was also significantly associated with diabetes (p < 0.001), with Indian and Malay ethnicity being associated with of the presence of diabetes, compared to Chinese ethnicity (Table 1).

Association of DM (Diabetes Mellitus) and outcomes

Compared to patients without diabetes, patients with diabetes tended to get postoperative hyperglycaemia (88.3% vs 56.0%; p < 0.001) and postoperative infection (2.8% vs 1.3%; p = 0.005) (Table 2).

Table 2.

Outcomes of patients with and without diabetes after cardiac surgery.

Perioperative variables	All (n = 2831)	Patients with diabetes (n = 1321)	Patients without diabetes (n = 1510)	p-value
Outcomes and complications
Postoperative hyperglycaemia %(n)	71.1 (2012)	88.3 (1166)	56.0 (846)	<0.001
Atrial fibrillation %(n)	24.4 (691)	23.5 (310)	25.2 (381)	0.271
Cardiac arrhythmia %(n)	21.2 (599)	21.2 (280)	21.2 (319)	0.969
AKI %(n)	30.6 (865)	36.4 (479)	25.6 (286)	<0.001
New need for dialysis %(n)	3.4 (96)	4.9 (65)	2.1 (31)	<0.001
Postoperative infection %(n)	2.0 (57)	2.8 (37)	1.3 (20)	0.005
Mortality in 30 days %(n)	0.6 (16)	0.9 (11)	0.3 (5)	0.073
ICU mortality %(n)	1.6 (44)	1.4 (18)	1.7 (26)	0.441
Readmission to ICU %(n)	4.6 (130)	6.0 (79)	3.4 (51)	0.001
Length of stay in ICU, mean ± SD	47.5 ± 105.6	52.6 ± 115.8	43.1 ± 95.7	0.017
Length of hospitalisation, mean ± SD	11.5 ± 12.1	12.5 ± 13.3	10.7 ± 10.8	<0.001
Highest 24 h glucose (mmol/L), mean ± SD	11.3 ± 2.7	12.4 ± 2.8	10.4 ± 2.2	<0.001
Highest 48 h glucose (mmol/L), mean ± SD	10.5 ± 3.3	12.2 ± 3.1	8.4 ± 2.1	<0.001

SD: standard deviation; ICU: intensive care unit; AKI: acute kidney injury.

Patients with diabetes also had poorer renal outcomes with diabetes being significantly associated with AKI (36.4% vs 25.6%; p < 0.001) and new need for dialysis (4.9% vs 2.1%; p < 0.001) (Table 2).

Diabetes patients also tended to have significantly higher rates of readmission to ICU (6.0% vs 3.4%; <0.001), longer stay in ICU (52.6 ± 115.8 days vs 43.1 ± 95.7 days; p = 0.017), and longer hospitalisation (12.5 ± 13.3 vs 10.7 ± 10.8; p < 0.001) (Table 2).

From multivariate analysis, diabetes mellitus was found to significantly increase the odds of ICU readmission by 1.70 (95% CI 1.171–2.480, p = 0.005), postoperative infection by 1.73 (95% CI 1.003–2.976, p = 0.049), AKI by 1.36 (95% CI 1.137–1.626, p = 0.001), postoperative hyperglycaemia by 6.00 (95% CI 4.893–7.348, p < 0.001), and the new need for dialysis by 1.71 (95% 1.086–5.360, p = 0.021) (Table 3).

Table 3.

Perioperative outcome variables significantly associated with diabetes mellitus after multivariable logistic regression analysis for each outcome.

Outcome variable	OR	95% CI	p-value
	Diabetes mellitus		p-value
ICU readmission	1.70	1.171–2.480	0.005
Postoperative infection	1.73	1.003–2.976	0.049
New dialysis	1.71	1.086–5.360	0.021
AKI	1.36	1.137–1.626	0.001
Postoperative hyperglycaemia	6.00	4.893–7.348	<0.001

OR: odds ratio; CI: confidence interval; AKI: acute kidney injury; ICU: intensive care unit.

Subsequent subgroup analysis of patients with diabetes revealed that compared to those who were only on oral hypoglycaemic agents (OHGA), those who were on insulin regimes tended to be predisposed to poorer outcomes of postoperative hyperglycaemia (93.4% vs 87.0%, p < 0.001), AKI (49.0% vs 36.3%, p < 0.001), new need for dialysis (10.2% vs 3.7%, p < 0.001), ICU mortality (3.1% vs 0.9%, p = 0.006), and longer hospitalisation stay (15.3% vs 11.8%, p < 0.001) (Table 4).

Table 4.

Patient outcomes of diabetes patient treated with insulin versus those on OHGA only.

Perioperative variables	Patients with diabetes (n = 1321)	Diabetes patients on insulin (n = 258)	Diabetes patients not on insulin (n = 1063)	p-value
Preoperative baseline variables
Female %(n)	21.3 (282)	31.9 (81)	18.7 (199)	<0.001
Weight, kg, mean ± SD	67.6 ± 13.1	66.9 ± 13.9	67.7 ± 12.9	0.356
Height, cm, mean ± SD	162.8 ± 8.1	161.7 ± 8.1	163.1 ± 8.1	0.017
BMI, kg/m², mean ± SD	25.4 ± 4.2	25.5 ± 4.8	25.4 ± 4.0	0.619
Age, years, mean ± SD	60.6 ± 9.0	60.6 ± 9.1	60.6 ± 9.0	0.956
Chinese %(n)	65.6 (867)	63.4 (161)	66.1 (703)	0.351
Malay %(n)	17.0 (225)	16.1 (41)	17.2 (183)
Indian %(n)	17.3 (229)	20.5 (52)	16.7 (177)
Preoperative renal impairment %(n)	16.9 (223)	31.5 (80)	13.5 (143)	<0.001
Dialysis %(n)	5.9 (78)	11.8 (30)	4.5 (48)	<0.001
Serum creatinine >200 µmol/L %(n)	7.0 (92)	15.4 (39)	5.0 (53)	<0.001
LVEF <30 %(n)	11.0 (145)	12.2 (31)	10.6 (113)	0.502
Previous cardiac surgery %(n)	2.0 (26)	3.5 (9)	1.6 (17)	0.045
Hypertension %(n)	88.5 (1165)	89.8 (228)	88.1 (937)	0.469
Arrhythmia %(n)	7.5 (99)	5.9 (15)	7.9 (84)	0.278
Atrial fibrillation %(n)	5.7 (75)	2.8 (7)	6.4 (68)	0.024
HbA1c (%), mean ± SD	7.6 ± 2.1	8.3 ± 2.0	7.4 ± 2.1	<0.001
Preoperative glucose (mmol/L), mean ± SD	8.7 ± 3.8	9.7 ± 4.2	8.4 ± 3.7	<0.001
Outcomes and complications
Postoperative hyperglycaemia %(n)	88.3 (1166)	93.4 (241)	87.0 (925)	<0.001
Atrial fibrillation %(n)	23.5 (310)	19.7 (50)	24.5 (260)	0.107
Cardiac arrhythmia %(n)	21.2 (280)	23.2 (59)	20.8 (221)	0.402
AKI %(n)	36.4 (479)	49.0 (124)	36.3 (477)	<0.001
New need for dialysis %(n)	4.9 (65)	10.2 (26)	3.7 (39)	<0.001
Postoperative infection %(n)	2.8 (37)	4.3 (11)	2.5 (26)	0.105
Mortality in 30 days %(n)	0.9 (11)	0.9 (2)	0.9 (9)	0.959
ICU mortality %(n)	1.4 (18)	3.1 (8)	0.9 (10)	0.006
Readmission to ICU %(n)	6.0 (79)	7.9 (20)	5.6 (59)	0.168
Length of stay in ICU, mean ± SD	52.6 ± 115.8	64.5 ± 143.4	49.7 ± 108.2	0.069
Length of hospitalisation, mean ± SD	12.5 ± 13.3	15.3 ± 17.7	11.8 ± 11.9	<0.001
Highest 24 h glucose(mmol/L), mean ± SD	12.4 ± 2.8	13.0 ± 3.0	12.3 ± 2.8	<0.001
Highest 48 h glucose(mmol/L), mean ± SD	12.2 ± 3.1	13.4 ± 3.4	11.9 ± 3.0	<0.001

SD: standard deviation; BMI: body mass index; LVEF: left ventricular ejection fraction; ICU: intensive care unit; AKI: acute kidney injury; OHGA: oral hypoglycaemic agent; HbAlc: haemoglobin A1c.

Discussion

Overall prevalence of DM

The overall prevalence of DM in our Southeast Asian cohort undergoing cardiac surgery was 46.7%. This is significantly higher than that reported in Western populations (12%–38%),¹² supporting findings that Asia is the major site of a rapidly emerging diabetes epidemic.¹³ This further validates the need of our study to determine whether diabetes is associated with poorer cardiac surgery outcomes among Southeast Asian patients.

Association of DM and poorer renal outcomes

Our findings show that diabetes patients were shown to be 1.36 times more likely than patients without diabetes to develop AKI. Previous studies have shown that diabetes patients are at increased risk of poorer renal outcomes, even after adjusting for numerous known risk factors, including chronic kidney disease.¹⁴ Similar Western cohort studies have reported that surgical patients with diabetes were about 1.40 times more likely to have renal insufficiency and AKI after surgery as compared to those without diabetes,^3,15 similar to our findings.

We also found that diabetes patients were over 1.71 times more likely to need new dialysis after surgery as compared to those without diabetes. The poorer renal outcomes (AKI, new need for dialysis) associated with diabetes are likely because patients with diabetes undergoing cardiac surgery have a greater inflammatory and oxidative stress response with significantly higher levels of cortisol, hsCRP, interleukin-6, and oxidative stress markers during the postoperative period, which is in turn associated with renal damage and poorer postoperative renal outcomes.¹⁶

In a long-term prospective observational study of 4273 UK Prospective Diabetes Study (UKPDS), participants followed for a median of 18 years found similar results.¹⁷ The study found that there were no differences in the rates of stroke or microvascular disease between the three ethnic groups in the study. This corroborates the findings in this study that the postoperative risk of developing complications among Asian diabetes patients is similar to that of Western diabetes patients, as compared to patients without diabetes.

Association of DM and postoperative infections and hyperglycaemia

We found patients with diabetes to be almost twice as likely to develop postoperative infections than those without diabetes.^18,19 Similarly, other Western cohort studies have also found diabetes patients to be 2.03 more likely to develop postoperative infections, suggesting that patients with diabetes are twice as likely as those without diabetes to develop postoperative hyperglycaemia regardless of whether they are of Asian or Caucasian origin. Expectedly, the presence of diabetes is also strongly associated with the onset of postoperative hyperglycaemia (increasing its risk of onset by six times), which has been shown to lead to poorer outcomes in cardiac surgery patients.²⁰ Such post-surgical hyperglycaemia abolishes intrinsic myocardial protective mechanisms such as ischemic preconditioning²¹ and causes both macrovascular and microvascular complications, as well as an increased risk of cardiovascular mortality.²²

Furthermore, this increased rate of postoperative hyperglycaemia in diabetes patients has been associated with deep sternal wound infections, further worsening prognosis and outcomes for diabetes patients undergoing cardiac surgery.^6,23 This is because the hyperglycaemia among diabetes patients impairs their immune function and the inflammatory response which are critical for postoperative healing.¹⁹

Association of DM and poorer hospitalisation outcomes

We also found diabetes patients to have a higher risk of being readmitted to the ICU compared to those without diabetes (OR 1.70, 95% CI 1.171–2.480, p = 0.005). Patients with diabetes also tended to have significantly longer ICU stay (52.6 vs 43.1, p = 0.017) and longer hospitalisation (12.5 vs 10.7, p < 0.001), likely due to the presence of complications and comorbidities associated with diabetes, increasing the cost of treatment and burden on healthcare.

Diabetes patients treated with insulin

Subsequent subgroup analysis of patients with diabetes revealed that compared to those who were only on oral hypoglycaemic agents (OHGAs), those who were on insulin tended to be predisposed to poorer outcomes of postoperative hyperglycaemia (93.4% vs 87.0%, p < 0.001), AKI (49.0% vs 36.3%, p < 0.001), new need for dialysis (10.2% vs 3.7%, p < 0.001), ICU mortality (3.1% vs 0.9%, p = 0.006), and longer hospitalisation stay (15.3% vs 11.8%, p < 0.001). Zarling et al.²⁴ also showed that patients on insulin therapy tended to be 1.89 times more likely to require Discharge to an Extended-Care Facility after surgery, corroborating our results that patients with diabetes who are on insulin therapy are more likely to have poorer outcome and more postoperative complications than their counterparts who have not been started on regular insulin regimes.

We postulate this is because requiring preoperative insulin treatment is a surrogate marker of poorer diabetes control and the presence of complications of diabetes. Patients with diabetes who were on insulin tended to have higher HbA1c (8.3% vs 7.4%, p < 0.001), higher preoperative glucose levels (9.7% vs 8.4%, p < 0.001), preoperative renal impairment (31.5% vs 13.5%, p < 0.001), serum creatinine >200 µmol/L (15.4% vs 5.0%, p < 0.001) and be on dialysis (11.8% vs 13.5%, p < 0.001) as compared to their counterparts with diabetes who were not on insulin. This is despite the fact that there was no significant difference between the subgroups in terms of baselines patient demographics such as weight, BMI, age, or ethnicity. Thus, it is clear that patients with diabetes who are on insulin therapy are likely to have more poorly controlled diabetes (and associated complications) and consequently worse outcomes after cardiac surgery than those not requiring insulin therapy (such as those only on OHGA).

Conclusion

In conclusion, diabetes is associated with increased risk for renal dysfunction, hyperglycaemia and infection after cardiac surgery as compared to patients without diabetes, to a similar extent to findings in Western populations. As such, patients with diabetes undergoing cardiac surgical interventions should receive special attention during the perioperative period to prevent the identified adverse effects, thereby improving postoperative outcomes of patients with diabetes.

Footnotes

Declaration of conflicting interests

The author(s) declared no potential conflicts of interest with respect to the research, authorship and/or publication of this article.

Funding

The author(s) received no financial support for the research, authorship and/or publication of this article.

ORCID iD

Sophia Tsong Huey Chew

References

Bank

IEM

Gijsberts

Teng

, et al. Prevalence and clinical significance of diabetes in Asian versus White patients with heart failure. JACC Heart Fail 2017; 5: 14–24.

Tan

KHX

Barr

ELM

Koshkina

, et al. Diabetes mellitus prevalence is increasing in South Asians but is stable in Chinese living in Singapore and Mauritius. J Diabetes 2017; 9: 855–864.

Bucerius

Gummert

Walther

, et al. Impact of diabetes mellitus on cardiac surgery outcome. Thorac Cardiovasc Surg 2003; 51: 11–16.

Gum

O’Keefe

Borkon

, et al. Bypass surgery versus coronary angioplasty for revascularization of treated diabetic patients. Circulation 1997; 96: II-7–II-10, http://www.ncbi.nlm.nih.gov/pubmed/9386067 (accessed 28 December 2016).

Lazar

McDonnell

Chipkin

, et al. The Society of Thoracic Surgeons practice guideline series: blood glucose management during adult cardiac surgery. Ann Thorac Surg 2009; 87: 663–669.

Garg

Grover

McGurk

, et al. Predictors of hyperglycemia after cardiac surgery in nondiabetic patients. J Thorac Cardiovasc Surg 2013; 145: 1083–1087.

Teo

Wang

, et al. GFR estimating equations in a multiethnic Asian population. Am J Kidney Dis 2011; 58: 56–63.

Summary of recommendation statements. Kidney Int Suppl 2011; 3: 5–14.

Das

Afsana

Elahi

, et al. Renal insufficiency among urban populations in Bangladesh: a decade of laboratory-based observations. PLoS ONE 2019; 14: e0214568.

10.

Whelton

Carey

Aronow

, et al. 2017 ACC/AHA/AAPA/ABC/ACPM/AGS/APhA/ASH/ASPC/NMA/PCNA guideline for the prevention, detection, evaluation, and management of high blood pressure in adults: executive summary: a report of the American College of Cardiology/American Heart Association Task Force on clinical practice guidelines. Hypertension 2018; 71: 1269–1324.

11.

Clark

RE.

Definitions of terms of the society of thoracic surgeons national cardiac surgery database. Ann Thorac Surg 1994; 58: 271–273.

12.

Haffner

SM.

Epidemiology of type 2 diabetes: risk factors. Diabetes Care 1998; 21: C3–C6, http://www.ncbi.nlm.nih.gov/pubmed/9850478 (accessed 31 December 2016).

13.

Ramachandran

Snehalatha

Diabetes in Asia. Lancet 2010; 375: 408–418.

14.

Girman

Kou

Brodovicz

, et al. Risk of acute renal failure in patients with type 2 diabetes mellitus. Diabet Med 2012; 29: 614–621.

15.

Kheterpal

Tremper

Heung

, et al. Development and validation of an acute kidney injury risk index for patients undergoing general surgery. Anesthesiology 2009; 110: 505–515.

16.

Reyes-Umpierrez

Davis

Cardona

, et al. Inflammation and oxidative stress in cardiac surgery patients treated to intensive vs. conservative glucose targets. J Clin Endocrinol Metab 2016; 102: jc20163197.

17.

Davis

TME

Coleman

Holman

. UKPDS Group. Ethnicity and long-term vascular outcomes in type 2 diabetes: a prospective observational study (UKPDS 83). Diabet Med 2014; 31: 200–207.

18.

Martin

Kaye

Knott

, et al. Diabetes and risk of surgical site infection: a systematic review and meta-analysis. Infect Control Hosp Epidemiol 2016; 37: 88–99.

19.

Aragon

Ring

Covelli

The influence of diabetes mellitus on postoperative infections. Crit Care Nurs Clin North Am 2003; 15: 125–135, http://www.ncbi.nlm.nih.gov/pubmed/12597048 (accessed 31 August 2017).

20.

Moorthy

Sim

Liu

, et al. Association between ethnicity and postoperative hyperglycemia in a Southeast Asian population undergoing cardiac surgery. J Cardiothorac Vasc Anesth 2019; 33: 388–393.

21.

Duncan

AE.

Hyperglycemia and perioperative glucose management. Curr Pharm Des 2012; 18: 6195–6203.

22.

Reddy

Abraham

Sundaram

, et al. Is there a genetic predisposition to new-onset diabetes after kidney transplantation? Saudi J Kidney Dis Transpl 2015; 26: 1113–1120.

23.

Polito

Thiagarajan

Laussen

, et al. Association between intraoperative and early postoperative glucose levels and adverse outcomes after complex congenital heart surgery. Circulation 2008; 118: 2235–2242.

24.

Zarling

Sikora-Klak

Bergum

, et al. How do preoperative medications influence outcomes after total joint arthroplasty? J Arthroplasty 2017; 32: S259–S262.