To the Editor,

COVID-19 critical illness in pregnant patients is associated with an increased likelihood of Cesarean delivery (CD) (72–84%).1, 2 We describe the general anesthetic technique for CD and postoperative outcomes of a parturient with COVID-19 acute respiratory distress syndrome (ARDS). Written informed consent for publication was obtained.

A 31-yr-old G3P2 at 26 weeks gestation was intubated in a community hospital for severe COVID-19 ARDS (caused by the Beta and Alpha variant of SARS-CoV-2) associated with fever and hypoxemic respiratory failure (Figure). Upon transfer to Mount Sinai Hospital (Toronto, ON, Canada), the patient received lung protective ventilation in the intensive care unit (ICU) using a pressure control mode. Initial parameters were Pinsp 16 cm H2O and positive end-expiratory pressure (PEEP) 10 cm H2O, targeting a tidal volume of 400 mL. Deterioration in her blood gas and PaO2/FiO2 ratio (pH 7.26; PaCO2 47 cm H2O; PaO2 96 cm H2O; FiO2 1.0) prompted paralysis with rocuronium (20 μg·kg-1·min-1) and proning. This was discontinued after 48 hr because of sustained improvement in her oxygenation and PaO2/FiO2 ratio (PaO2 118 cm H2O; FiO2 0.4). In the ICU, fetal status was monitored with daily nonstress testing and biophysical profile. Delivery was delayed given her stable results to allow for further fetal maturation. Ten days later, the patient developed septic shock. Increasing ventilatory requirements (Pinsp 25 cm H2O; PEEP 5 cm H2O; FiO2 1.0), hypoxia (SpO2 79%), and new infiltrates on chest x-ray prompted a diagnosis of ventilator-associated pneumonia (VAP). Her mean arterial pressure was maintained > 65 mm Hg with norepinephrine (0.02 μg·kg-1·min-1) and vasopressin (0.04 IU·min-1). Piperacillin-tazobactam was started. The patient was repositioned prone, with no change in SpO2. Inhaled nitric oxide was administered at 40 ppm, which improved the SpO2 to 93%. The patient’s hemoglobin fell from 97 g·L-1 to 68 g·L-1 without obvious blood loss. She suffered an acute renal injury without requiring renal replacement therapy. On day 20, a multidisciplinary decision was made to perform urgent CD because of preterminal fetal ultrasound.

Figure
figure ure

Chest x-ray. A) Day 1 of admission. Significant bilateral pulmonary opacification, in keeping with COVID-19 pneumonia. Blunting of the left costophrenic angle with no large pleural effusion present. B) Day 14 of admission. Pre-Cesarean delivery chest x-ray. Increasing bilateral opacities associated with presumptive diagnosis of ventilator-associated pneumonia superimposed on ARDS. C) Day 38 of admission. Interval improvement of bilateral chest opacities following decannulation from extracorporeal membrane oxygenation.

Full size image

Cesarean delivery was performed at 28+6 weeks gestation. The patient was maintained on ICU sedation: propofol 60 μg·kg-1·min-1, fentanyl 250 μg·hr-1, midazolam 10 mg·hr-1, and vasopressors. The patient was paralyzed with rocuronium 50 mg and ventilated with pressure-control ventilation using a PEEP of 15 cm H2O. She developed uterine atony and postpartum hemorrhage (PPH) of 1,500 mL requiring four units of packed red blood cells and 1,000 mL of Ringer’s lactate. Fluid balance was 500 mL positive. Uterotonic management included oxytocin (5U slow iv × 4 and 40 IU/L infusion at 125 mL·min-1), ergometrine (250 μg slow iv × 1 and im × 1), and carboprost tromethamine (250 μg intramyometrial). Calcium chloride 1 g and tranexamic acid 1 g twice were also administered. The fetus was delivered flaccid and apneic with APGAR scores of 2, 5, and 8 at one, five, and ten minutes, respectively. The patient’s uterine atony had resolved by the end of the surgery.

Postpartum, the patient’s respiratory status remained critical, likely from a combination of ARDS, VAP, and PPH-related fluid redistribution. Her ventilatory parameters were precarious (Pinsp 24 cm H2O; PEEP 10 cm H2O; FiO2 0.9–1.0); driving pressures above 20 cm H2O were required. Intermittent proning, paralysis, diuresis and escalation of her antimicrobial therapy was undertaken, with little clinical improvement. The patient was subsequently trialled on high-frequency oscillatory ventilation before being transferred for extracorporeal membrane oxygenation (ECMO) and tracheostomy on postoperative day 16. She was decannulated from ECMO after ten days. Her subsequent recovery was complicated by sporadic bacteremia and slow weaning of respiratory support. The patient was discharged from the ICU on postoperative day 26, and from the hospital after another month.

Parturients with COVID-19 ARDS present a unique challenge, combining considerations for ARDS, multisystem critical illness, and concurrent pregnancy. Severe hypoxemia and precipitous clinical decline served as the basis for mechanical ventilation and managing this case under general anesthesia; however, delivery did not improve her respiratory status.

Lung protective ventilation improves outcomes in patients with ARDS.3 Plateau pressures < 30 cm H2O and driving pressures < 15 cm H2O should be targeted. High PEEP (≥ 15 cm H2O) was employed for hypoxemia. Alveolar recruitment and improved oxygenation with higher PEEP should be weighed against the hemodynamic consequences of reduced cardiac preload.

We support the use of norepinephrine as the first-line agent in the treatment of hypotension for critically ill obstetric patients. Overzealous fluid and blood product administration can contribute to pulmonary edema in ARDS, which is likely to be exacerbated by auto-transfusion after delivery. This may have contributed to our patient’s worsening respiratory status post-CD. A restrictive fluid strategy and blood transfusion target of > 70 g·L-1 may be optimal, with vasopressors as required.4 Uterotonic agents should be used judiciously because of their hemodynamic and respiratory effects. Hemabate and ergometrine increase pulmonary vascular resistance and may cause pulmonary edema. Hemabate can also cause bronchospasm impairing ventilation and oxygenation.5

An intraoperative and postoperative decline in respiratory status should be anticipated in COVID-19 parturients with ARDS. Extracorporeal membrane oxygenation should be considered when oxygenation is inadequate despite mechanical ventilation, optimized FiO2, and prone positioning.