The haemodynamic effects of phenylephrine after cardiac surgery

Phenylephrine increases systemic‐ and pulmonary resistances and therefore may increase blood pressures at the expense of blood flow. Cardio‐pulmonary bypass alters vasoreactivity and many patients exhibit chronotropic insufficiency after cardiac surgery. We aimed to describe the haemodynamic effects of phenylephrine infusion after cardiac surgery.


| INTRODUCTION
Phenylephrine is an intravenous vasoconstrictor widely used in anaesthesia to treat systemic hypotension resulting from vasodilatation and consequent decrease in systemic vascular resistance (SVR). Phenylephrine counteracts vasodilatation predominantly by activating vascular smooth muscle α1-receptors, causing higher intracellular calcium concentration in myocytes by both release of stored calcium within the myocyte and influx of extracellular calcium. 1 Vasodilatation and vasoplegia are especially common during and after cardiac surgery due to the inflammatory effects of cardio-pulmonary bypass (CPB) 2 and the subsequent ischaemia-reperfusion injury. A recently published study of an American cohort of 104,963 patients revealed that 84% of noncomplex cardiac surgery patients received vasoconstrictors on the day of surgery, and during the entire hospitalisation phenylephrine was the most common vasoactive drug infused with 76.6% receiving it. 3 Anaesthesiologists may be hesitant to use phenylephrine as it increases systemic blood pressure, with only little or negative effect on organ flow, 4 thus causing seemingly favourable monitor values at the clinically invisible expense of tissue oxygenation. 5 Several studies have shown that phenylephrine decreases cardiac output (CO) [6][7][8] with concomitant increase in right ventricular pressure, consequent right ventricular dilation, 9 and decrease in heart rate.
Conversely, norepinephrine may preserve CO by maintaining preload and enhancing contractility. 10 Further, Kwak et al 7 found that, in patients with pre-existing pulmonary hypertension, phenylephrine was less reliable in increasing mean systemic arterial pressure (MAP) than norepinephrine, but phenylephrine did not significantly change the ratio between pulmonary vascular resistance (PVR) and SVR. This ratio is clinically interesting, as a change in the PVR/SVR ratio is a reflection of a more selective contraction of either the pulmonary-or systemic vascular bed.
Recent studies in hypotensive surgical patients 11,12 have brought additional perspective. In surgical settings with hypotension, a single bolus phenylephrine increased CO if pulse pressure variation (PPV) was high. Thus, the effects of phenylephrine are dependent on the individual hearts position on the Frank-Starling relationship.
Studies on phenylephrine have been conducted either outside cardiac surgery or prior to CPB. CPB alters smooth muscle intracellular calcium homeostasis and dephosphorylates myosin and may cause desensitisation of α1-receptors. 2 Along with an altered vasoreactivity, patients after CBP rely on epicardial pacing, which reduces the heart rate reactivity to vasoconstrictors. Hence, the circulatory effects of vasoactive medication found prior to CBP may not be applicable after CPB and the effects of phenylephrine after CPB are not well-described.
We aimed to describe the haemodynamic effects of phenylephrine infusion after low-risk cardiac surgery. Our main hypothesis was that, in haemodynamically stable patients, the PVR/SVR ratio would remain unchanged in response to a phenylephrine infusion when targeting a 20 mmHg increase in MAP.

| METHODS
This prospective study approved by the Regional Ethical Committee of the Central Denmark Region (identifier 1-10-72-40-20) on 5 June 2020 and it was conducted in accordance with the Helsinki II declaration. All patients gave written, informed consent prior to inclusion in the study. The study protocol was registered prior to patient enrolment at clinicaltrial.gov (identifier NCT04419662) and constitutes the final intervention in a study in which renal ultrasound measures have been published. 13 This manuscript adheres to the applicable TREND guidelines. Anaesthesia, CPB and postoperative care were conducted as previously described. 13 As part of routine monitoring, a 7.5 F thermodilution pulmonary artery catheter was inserted after anaesthesia induction and coupled to a Vigilance II monitor (Edwards Lifesciences, CA, USA) for continuous CO measurements. After surgery, sedated patients were transferred to the post-operative care unit and mechanically ventilated with an Evita Infinity V500 (Dräger, Lübeck, Germany). Tidal volume (6-8 mL/kg) (ideal weight) and the level of positive-end expiratory pressure (5-7 cmH 2 O) were determined by the attending physician. Norepinephrine was infused to maintain a MAP of a minimum of 65 mmHg.
The study intervention was commenced approximately 90-120 min after surgery. Patients were in haemodynamic steady state as confirmed by stable blood pressures, requirements for any vasoac- + min(PP))/2) and defined as the average of the two least extreme values calculated from the four cycles. Tidal volume was not indexed to weight and often below 8 mL/kg, defined in the literature as the minimally acceptable tidal volume for optimal preload responsiveness prediction. 14 Therefore, we constructed a tidal volume corrected PPV representing the PPV expected at tidal volumes of 10 mL/kg, as the relationship between tidal volume and PPV is close to proportional. 15 Tidal-volume corrected PPV was calculated as ideal weight Â 10 mL kg observed tidal volume : where ideal weight was estimated as the weight corresponding to a body mass index of 22 kg/m 2 . 16 Transoesophageal echocardiography was conducted with 6VT probe attached to Vivid S70/E9 ultrasound systems (GE Healthcare, Horten, Norway). All endpoints were achieved from the mid- The primary endpoint was the PVR/SVR ratio. Secondary endpoints were the other above-mentioned invasive pressures and echocardiographic indices.

| Statistical analyses
We expected the PVR/SVR ratio to be around 0.078 with an estimated standard deviation of 30% from previously published data. 8 Including 30 patients would enable detection of a 15% change in the PVR/SVR ratio, deemed clinically relevant (α = 0.05, 1 À β = 0.8,  Table 1.  Table 3 for details on all echocardiographic indices.   7 The PVR/SVR ratio did not change in our study which is consistent with findings comparing phenylephrine to norepinephrine. 7 The increased PVR challenges the right ventricular contractile reserve which may be of concern after cardiac surgery and the inherent myocardial stunning of CPB. However, we did not see right ventricular dilatation secondary to the increase in afterload and, as substantiated by the overall increase in CO, right ventriculo-arterial coupling was intact. Therefore, the increase in PVR does not discourage the use of phenylephrine after cardiac surgery in low-risk patients.

| DISCUSSION
In patients of increased risk of right ventricular failure and vasodilatation, vasopressin seems to be a better choice due to its minimal effect on PVR. 18

| Limitations
Phenylephrine administration is often accompanied by reflex bradycardia 20 which in the current study was not possible due to pacing. It can therefore be claimed that we overrode normal pharmacodynam-

| CONCLUSION
In haemodynamically stable patients after cardiac surgery, phenylephrine increased PVR and SVR, but did not change the PVR/SVR ratio.
Phenylephrine increased biventricular filling pressures and consequently increased CO. These findings do not discourage the use of phenylephrine after low-risk cardiac surgery in haemodynamically stable patients.

AUTHOR CONTRIBUTIONS
Peter Juhl-Olsen concieved the study, performed the study, did statis-