Myocardial blood flow under general anaesthesia with sevoflurane in type 2 diabetic patients: a pilot study

The local Human Subjects Ethics Committee of the VU University Medical Center in Amsterdam, the Netherlands, approved this study (ID 2008/304). Subsequent patients were included after written informed consent. We prospectively studied 6 patients with type 2 (non-insulin dependent) diabetes mellitus and 8 healthy control subjects, all scheduled for non-cardiac surgery under general anaesthesia. Six out of 8 healthy volunteers were selected based on sex from a previously published study by Bulte et al.[17]. The diabetic subjects had haemoglobin A1c levels of 6.8 ± 0.8% (range 5.5 – 7.6%) and fasting glucose levels of 8.5 ± 1.5 mmol/l (range 6.7 – 10.9 mmol/l). Mean duration of type 2 diabetes was 7.7 ± 5 year (range 3 – 15 years), diabetic patients only received oral antidiabetic therapy. Exclusion criteria for participation included age < 18 years, allergy to echocardiographic contrast agents or adenosine, previous history of coronary artery disease, chronic obstructive pulmonary disease and use of beta-adrenergic blocking agents.

All participants made an extra visit to our hospital for screening, autonomic function testing and baseline myocardial blood flow measurements. On the day of surgery the MBF measurements were repeated after the induction of sevoflurane anaesthesia but before the start of the surgical procedure.

As previously described, transthoracic myocardial contrast echocardiography (MCE) was performed using an iE33 ultrasound scanner equipped with a S5 – 1 transducer (Philips Medical Systems, Best, The Netherlands)[18]. A contrast agent consisting of microbubbles filled with sulphur hexafluoride with a mean diameter of 2,5 μm was used (Sonovue, Bracco Imaging, Milan, Italy) and continuously infused using a specific syringe pump (VuEject, Bracco SA, Switzerland). After two minutes of microbubble infusion, baseline perfusion images were acquired from apical 4-, 2- and 3-chamber views. Subsequently, hyperaemia was induced by a continuous infusion of 0.14 mg kg^-1 min^-1 adenosine. Finally, sympathetic stimulation was triggered by the cold pressor test (CPT) by immersing the hand of the patient in ice water (2 – 4°C) for 3 minutes. Perfusion sequences were recorded and consisted of 10 cardiac cycles of low acoustic power (mechanical index [MI] 0.17) imaging for microbubble detection followed by a burst of high acoustic power (MI 0.64) for microbubble destruction. Subsequently, 20 cardiac cycles were recorded with low MI imaging at a frame rate of 18 Hz to allow contrast replenishment in the myocardium. All data were stored for offline analysis.

All subjects underwent a set of standard autonomic function tests to screen for cardiovascular autonomic neuropathy. First, subjects are placed in the supine position and heart rate (R – R intervals) was recorded for 5 min during spontaneous breathing. Subsequently, autonomically induced variations in R – R intervals to changes in blood pressure were recorded during one minute of deep breathing, during the Valsalva manoeuvre (Valsalva ratio) and quick stand test. R – R intervals and blood pressure were continuously recorded using a non-invasive continuous finger arterial blood pressure measurement device with a sample rate of 200 Hz (Nexfin HD, BMEYE, Edwards Lifesciences, The Netherlands). Data was stored on a personal computer for further analysis using free available software (Kubios HRV version 2.0, University of Kuopio, Finland and Beatscope, BMEYE, Edwards Lifesciences, The Netherlands). During spontaneous breathing over 5 minutes, heart rate variability (HRV) was assessed by spectral analysis using fast Fourier transformation. This method divides the overall variability of a signal into its composing frequencies and provides insight into what extent a frequency contributes to the overall variability. The power spectrum of HRV consists of three peaks: the very-low-frequency band (<0.04 Hz), the low-frequency band (0.04 – 0.12 Hz) and the high-frequency band (0.12 – 0.40 Hz). The very-low-frequency fluctuations mediated primarily by the sympathetic system, the low-frequency fluctuations by both the sympathetic and parasympathetic system and the high-frequency fluctuations are under parasympathetic control.

From the recordings during deep breathing, maximum and minimum R – R intervals were extracted and expressed as a ratio, which in healthy subjects should be larger than 1.17. For the Valsalva manoeuvre, subjects exhaled forcibly through a manometer against a pressure of 40 mmHg for 15 s. The ratio of R – R intervals was calculated, being > 1.21 in healthy subjects. Assessment of heart rate response during quick standing, R – R intervals were measured at 15 and 30 beats after standing. In healthy subjects, the ratio of the longest R – R interval to the shortest R – R interval is > 1.04. To assess changes in blood pressure during standing, systolic blood pressure was measured in the resting, supine subject. Two minutes after rapid standing, the measurement was repeated. In healthy subjects the fall in systolic blood pressure after the test should be less than 10 mmHg. Cardiac autonomic neuropathy is defined as the presence of 3 or more abnormal test results among these 7 tests[19].

Heart rate, blood pressure and oxygen saturation were monitored throughout the protocol. All patients received midazolam 0.02 mg kg^-1 intravenously. Anaesthesia was induced by sevoflurane inhalation (AbbVie, Hoofddorp, The Netherlands) and maintained at an age-adjusted end-tidal concentration of 1.0 minimum alveolar concentration. A laryngeal mask airway was inserted and patients continued to breathe spontaneously without positive airway pressure during the study period. The surgical procedure started after MBF measurements were completed.

Semiautomated software providing manual region of interest (ROI) tracking, visualisation of perfusion calculations and dataset handling was used (PerfusionFitter, Department of Cardiology, Bern University Hospital). ROIs were drawn in end-systolic frames in the mid-inferoseptal and mid-anterolateral wall (apical 4-chamber view); in the mid-inferior and mid-anterior wall (apical 2-chamber view) and in the mid-inferolateral and mid-anteroseptal wall (apical 3-chamber view) of the myocardium. Using the volumetric model by Vogel et al., myocardial blood flow was quantified in ml min^-1 g^-1 from the underlying microvascular parameters: (1) the relative myocardial blood volume, which represents the blood volume in the capillary system, and (2) the capillary exchange rate, which provides an estimate of the exchange rate of erythrocytes within the region of interest[20]. Changes in blood pressure and heart rate during the interventions were monitored at regular intervals. Additional calculations were made to provide an estimate of myocardial oxygen consumption (rate-pressure product [RPP]; heart rate x systolic blood pressure) and coronary vascular resistance (CVR; dividing mean arterial blood pressure by MBF).

All data are presented as mean ± SD unless indicated otherwise. Baseline characteristics between controls and diabetics were compared using a Mann–Whitney U test. A Wilcoxon signed-rank test was used for within group comparisons of myocardial blood flow results (baseline versus sevoflurane). Changes in myocardial blood flow responses before and during sevoflurane administration were compared between controls and diabetics using a Mann–Whitney U test. Descriptive statistics were provided for hemodynamic data. A P < 0.05 was considered as statistically significant.

Table 1 shows baseline characteristics of the study population. Eight cardiovascular healthy males and six diabetic males were included in this pilot study. Type 2 diabetic patients were older than control subjects. No statistical difference in BMI is observed between groups. Fasting glucose and HbA1c-levels were higher in patients with type 2 diabetes. Further, total and LDL-cholesterol levels were lower in diabetics.

Clinical assessment of autonomic function in the control group revealed no abnormal responses to the tests (Table 2). In the diabetic group, one patient was identified with autonomic dysfunction based on three abnormal test results (heart rate response to deep breathing and standing, blood pressure response to standing). One patient had two abnormal responses (power spectrum in the low-frequency band and heart rate response to deep breathing), indicating borderline autonomic dysfunction.

At baseline, heart rate and RPP increased in both groups during adenosine infusion (Table 3). These responses were blunted in both groups during adenosine-induced hyperaemia under sevoflurane anaesthesia. The CPT at baseline increased SBP and RPP in diabetics and controls. Under sevoflurane anaesthesia, the CPT decreased SBP and RPP in healthy controls, while heart rate and RPP were decreased in diabetics.

Sevoflurane anaesthesia decreased MBF in diabetics but not in controls (P = 0.03). At baseline, adenosine-induced hyperaemia increased MBF in both groups compared to resting values. However, when adenosine was combined with sevoflurane anaesthesia, the MBF was significantly lower in both controls and diabetics when compared to baseline conditions (Figure 1). Differences in MBF response to adenosine before and after sevoflurane administration were larger in diabetic patients, this was however not statistically significant in this pilot group (P = 0.08). Myocardial blood flow parameters after the cold pressor test were not different between groups at baseline and during sevoflurane exposure.