Acute Respiratory Distress in the Operating Room and Prone Ventilation

Vishal Yajnik, MD, MS, Kathryn M. Breslin, MD, and Christa Riley, MD

> There have been many advances in the management of acute respiratory distress syndrome, a condition which Bellani et al, in the LUNG SAFE trial (Large Observational Study to Understand the Global Impact of Severe Acute Respiratory Failure), found represents up to 10.4% of intensive care unit admissions and 23.4% of patients requiring mechanical ventilation, with an unadjusted intensive care unit and hospital mortality of 35.3% and 40%, respectively. Studies have shown that prone positioning can improve oxygenation in patients who are mechanically ventilated for acute respiratory distress syndrome. This case report describes an example in which intraoperative prone positioning improved oxygenation in a patient after aspiration of gastric contents on induction of general anesthesia. (A&A Practice. 2019;12:19–21.)

There have been many advances in the management of acute respiratory distress syndrome, a condition which Bellani et al, in the LUNG SAFE trial (Large Observational Study to Understand the Global Impact of Severe Acute Respiratory Failure), found represents up to 10.4% of intensive care unit admissions and 23.4% of patients requiring mechanical ventilation, with an unadjusted intensive care unit and hospital mortality of 35.3% and 40%, respectively. Studies have shown that prone positioning can improve oxygenation in patients who are mechanically ventilated for acute respiratory distress syndrome. This article describes an example in which intraoperative prone positioning improved oxygenation in a patient after aspiration of gastric contents on induction of general anesthesia.

Diagnostic criteria for acute respiratory distress syndrome (ARDS) have improved as we have better elucidated its pathophysiology. There is now a broad range of treatment modalities, depending on the severity of the disease. Numerous studies have shown that prone positioning can improve oxygenation in patients requiring mechanical ventilation for this condition.2–5 We report the use of prone positioning to alleviate hypoxemia following intraoperative aspiration pneumonitis, which presented with many characteristics of ARDS. Our institutional written Health Insurance Portability and Accountability Act of 1996 authorization was obtained to publish this article.

A 59-year-old man with a medical history that included hypertension, diabetes mellitus, hyperlipidemia, and obesity (body mass index = 35) presented to another hospital with abdominal pain, nausea, vomiting, and constipation. He was found to have an incarcerated umbilical hernia with small-bowel obstruction, transferred to our institution for surgical management, and taken to the operating room. A rapid sequence induction was performed and the cricoid pressure was applied. The first attempt at intubation with direct laryngoscopy was unsuccessful despite a Cormack-Lehane grade I view due to an inability to pass the 8.0-mm endotracheal tube past the vocal cords. During the second attempt at intubation, a large volume of gastric contents filled the oropharynx. After suctioning, intubation was successful, with a Cormack-Lehane grade I view but using a smaller, 7.5-mm endotracheal tube that likely facilitated its passage. Oxygen saturation immediately postintubation ranged from 91% to 93%, despite Fio2 100%, but returned to a range of 96%–97% a few minutes thereafter. Bronchoscopy revealed a diffusely erythematous tracheobronchial tree, with brown-green contents that were suctioned from the mainstem bronchi. The arterial blood gas an hour after induction, when desaturation to the low 90s was noted, showed a pH of 7.38, Paco 48 mm Hg, Pao 22 73 mm Hg, HCO3 28.5 mmol/L, and positive end-expiratory pressure (PEEP) was increased to 10cm H2O. Surgery proceeded with primary repair of the incarcerated umbilical hernia, while the patient was maintained on 95%–100% Fio2, with volume control ventilation (6 mL/kg tidal volume, rate of 16 breaths per minute, and a PEEP of 8cm H2O), maintaining oxygen saturation between 87% and 94%, with normal peak and plateau airway pressures.

At the time of wound closure, the patient’s arterial oxygen saturation continued to decline, moving below 90%. The arterial blood gas at this time bore a pH of 7.38, Paco2 38 mm Hg, Pao2 45 mm Hg, HCO3 22.6 mmol/L. Recruitment breaths were delivered without improvement. An intensive care unit ventilator (Servo-I with NAVA, Maquet Critical Care AB, Solna, Sweden) was brought to the operating room to permit our use of bilevel ventilation settings, with high PEEP 25 cm H2O, low PEEP 12 cm H2O. A chest radiograph was obtained intraoperatively (Figure) to confirm appropriate positioning of the endotracheal tube and to exclude other potential causes of hypoxemia. A second bronchoscopic examination confirmed correct placement of the endotracheal tube and persistently erythematous, edematous airways bilaterally. Mucus plugs were not observed. As the patient’s hypoxemia persisted, he became progressively hypotensive. A femoral central venous catheter was placed through which norepinephrine was infused. His oxygen saturation further declined below 70%, and inhaled nitric oxide was initiated at 20ppm with minimal improvement in oxygen saturation. His bilevel settings were adjusted to high PEEP 28cm H2O and low PEEP 14 cm H2O, without improvement.

Given our patient’s refractory hypoxemia, we ultimately decided to manually place him in the prone position, using the assistance of operating room staff. His oxygen saturation increased to the 90%–94% range within a minute after repositioning him, and his arterial blood gas at that time bore a pH of 7.31, Paco2 47 mm Hg, Pao2 163 mm Hg, HCO3 23.9 mmol/L. He was transported to the surgical intensive care unit, where he remained in the prone position for 12 hours postoperatively. He demonstrated an improving Pao2/Fio2 (P/F) ratio, including an arterial blood gas of pH 7.36, Paco2 41 mm Hg, Pao2 276 mm Hg, and HCO3 23.4 mmol/L 1 hour later. He was extubated on the third postoperative day and discharged to a rehabilitation facility a week later, without any neurological deficits or cardiopulmonary sequelae from his complex hospitalization.

The Berlin definition of ARDS, most recently updated in 2012, requires (1) an onset of <7 days from purported insult, (2) a P/F ratio ≤300, (3) bilateral opacities consistent with pulmonary edema, detected on chest x-ray or computed tomography scan, and (4) that the condition cannot be fully explained by heart failure.5 Defining ARDS with modifications to the Berlin definition were shown by Riviello et al,6 especially in resource-deprived settings, to serve as a reasonable alternative to help with diagnosis. This was accomplished with measures such as Spo2/Fio2 ratios as an alternative for P/F ratios, no minimal PEEP requirement, and lung ultrasound. Notably, it was found that the Spo2/Fio2 ratio bore an 80% correlation to P/F ratios, increasing correlation to 90% when Spo2 was lower than 96%. The validity of this alternative has been noted via a nonlinear equation in multiple studies.7–9 In this patient’s case, this alternative, also known as the Kigali modification, proves useful because it is difficult to make a diagnosis of ARDS by Berlin definition criteria. Use of the modification allows the provider to recognize the severity of disease without these formal criteria and to accordingly guide therapeutic intervention.

Mechanical ventilation strategies for ARDS have centered on low-tidal volume ventilation to minimize alveolar overdistension—sometimes with concomitant permissive hypercapnia.3 The optimization of PEEP while maintaining an inspiratory plateau pressure of <30 cm H2O serves to maintain adequate oxygenation while decreasing the likelihood of barotrauma. This strategy of maintaining oxygenation and ventilation, while minimizing ventilator-induced lung injury, gives the diseased lungs time to heal from their initial insult. Ventilation-perfusion mismatching leads to impaired oxygenation. The dorsal lung, compressed by the patient’s body weight and ventral organs, receives preferential, gravity-dependent perfusion, while the ventral lung maintains more alveolar integrity and thus receives preferential ventilation with proportionately less perfusion.

Prone ventilation, which has been used to improve oxygenation and ventilation in ARDS patients since 1974,2 homogenizes ventilation in the diseased lung and promotes alveolar recruitment, thereby reducing ventilation-perfusion mismatch and ventilator-induced lung injury while improving oxygenation.10 In the prone position, the previously dependent lung continues to receive the majority of perfusion while its alveolar units open, while the newly dependent lung receives a minority of blood flow while its alveolar units start to slowly collapse.2 Moreover, the relief from cardiac and abdominal compression, and anatomically facilitated drainage of extravascular lung water, also contribute to the aforementioned improvement in lung aeration. The nature of lung injury typically seen in ARDS is typically heterogeneous and beyond excessive tidal volumes. PEEP that is implemented for therapeutic benefit can indeed prove damaging through overdistension of alveoli.3 Prone positioning may further aid in limiting this aspect of lung injury exacerbation by reducing regional hyperinflation and atelectasis while maintaining adequate aeration of the injured lungs. Interestingly, Cornejo et al11 found that damaging, cyclic alveolar recruitment/derecruitment was lowest when high PEEP and prone positioning were applied together.

Much of the early literature relating to prone ventilation in ARDS described improved oxygenation without reduced mortality. In the PROSEVA study (Prone Positioning for Severe Acute Respiratory Distress Syndrome),4 patients were randomly assigned to early (within 33 hours of intubation), high dose (17 consecutive hours) prone positioning if they met ARDS criteria. A large, statistically significant benefit was found in 28-day mortality (16% vs 33%; hazard ratio, 0.39; 95% CI, 0.25–0.63) and 90-day mortality (24% vs 41%; hazard ratio, 0.44; 95% CI, 0.29–0.67, respectively) as compared with patients ventilated in the supine position.4 While the study had limitations, it provided evidence that ARDS patient outcomes could be improved with an inexpensive intervention.

Our case demonstrates that prone positioning, not routinely considered in the operating room, quickly and effectively improved oxygenation in a patient with acute respiratory distress with features resembling ARDS. Our initial conservative treatment measures proved futile, and placing him in the prone position at the end of surgery resulted in a rapid improvement in his oxygenation. This institution of prone positioning likely aided in our patient’s good outcome, but it should be noted that he was kept in the prone position for only 12 hours, less than what Guérin et al4 noted to be of therapeutic benefit in the PROSEVA trial. Other therapies that were attempted included alveolar recruitment maneuvers, bilevel ventilation, and inhaled nitric oxide administration, none of which yielded notable improvement.

It should furthermore be noted that placing a patient in the prone position is neither simple nor without risks. One must be mindful of the possibility of unintended extubation or accidental removal of vascular catheters during the turning process. Maintaining prone positioning for a significant time period can also place the patient at risk for pressure and nerve injuries if prevention measures are not taken. At our institution, prone positioning is not uncommon, which helped with the efficient, team-based approach to transitioning from the supine to prone position. We suggest that institutions wishing to promote this ventilatory strategy, especially in the operating room, should develop a multidisciplinary protocol with clear patient selection criteria.

In conclusion, our case emphasizes the importance and value of early recognition of acute respiratory distress with hypoxemia and the utility of early prone positioning in the operating room.

Question 1

Diagnostic criteria for acute respiratory distress syndrome (ARDS) have improved since we have a better understanding of its:

Diagnostic criteria for acute respiratory distress syndrome (ARDS) have improved as we have better elucidated its pathophysiology.

Question 2

Numerous studies have shown that prone positioning can improve which of the following in patients requiring mechanical ventilation for ARDS?

Numerous studies have shown that prone positioning can improve oxygenation in patients requiring mechanical ventilation for this condition.

Question 3

Which of the following occurs after intraoperative aspiration pneumonitis, and is also characteristic of ARDS?

The authors reported the use of prone positioning to alleviate hypoxemia following intraoperative aspiration pneumonitis, which presented with many characteristics of ARDS.

Question 4

Much of the early literature relating to prone ventilation in ARDS described improved oxygenation without reduced:

Much of the early literature relating to prone ventilation in ARDS described improved oxygenation without reduced mortality.

Question 5

Prone positioning likely aided in the patient’s good outcome, but it should be noted that the patients were kept in the prone position for only 12 hours, less than what Guérin et al noted to be of therapeutic benefit in which trial?

Prone positioning likely aided in the patient’s good outcome, but it should be noted that he was kept in the prone position for only 12 hours, less than what Guérin et al noted to be of therapeutic benefit in the PROSEVA trial.

Question 6

A chest radiograph was obtained intraoperatively to confirm appropriate positioning of the endotracheal tube and to exclude other potential causes of:

A chest radiograph was obtained intraoperatively to confirm appropriate positioning of the endotracheal tube and to exclude other potential causes of hypoxemia.

Question 7

Other therapies during the PROSEVA trial that were attempted included alveolar recruitment maneuvers, bilevel ventilation, and inhaled administration of ______, none of which yielded notable improvement?

Other therapies during the PROSEVA trial that were attempted included alveolar recruitment maneuvers, bilevel ventilation, and inhaled administration of nitric oxygen, none of which yielded notable improvement.

Question 8

The nature of lung injury typically seen in ARDS in addition to excessive tidal volumes is due to:

The nature of lung injury typically seen in ARDS is typically heterogeneous and beyond excessive tidal volumes.