Definition and Recognition

Hypoxemia used to be a frequent occurrence during one-lung ventilation (OLV) before the use of routine fiberoscopy and modern inhaled anesthetic agents that have less of an inhibitory action on hypoxic vasoconstriction. Although the majority of cases requiring OLV now proceed without significant desaturation when lung isolation is achieved successfully, significant desaturation may still occur in 1% to 10% of patients even when maintained on a fraction of inspired oxygen (Fi o ₂ ) of 1.0. This case illustrates how hypoxia may still pose a significant challenge to the anesthesiologist. An understanding of the determinants of lung perfusion during OLV and a systematic approach to dealing with hypoxia as outlined in this chapter will provide the anesthesiologist with strategies to manage cases such as the one described.

In the absence of endotracheal tube or bronchial blocker malposition, hypoxia during OLV is caused primarily by shunting of blood through the nonventilated lung. The development of atelectasis in the ventilated lung further worsens this by creating additional areas of low V/Q. The blood flow from these nonventilated areas then mixes with that from the ventilated lung segments and causes a decrease in Pa o ₂ . Interestingly, no threshold saturation has been universally adopted as a safe lower limit during OLV; however, generally a saturation of 90% is often accepted, and in patients without significant comorbid disease, desaturations to the high 80s may be tolerated for short periods.

Optimal V/Q matching during OLV is achieved by maximizing perfusion to the dependent lung and maximizing pulmonary vascular resistance (PVR) in the nondependent lung. Many factors are at play in determining the proportion of pulmonary perfusion that flows through the nonventilated lung during OLV, and several are amenable to influence by the anesthesiologist ( Fig. 66.1 ). Lung volumes influence PVR in a hyperbolic manner; at extremes the PVR increases whereas at functional residual capacity (FRC) it is at its lowest, which is ideal for the dependent lung. During anesthesia, atelectasis tends to develop in the dependent lung because of mechanical compression in the lateral position and gas absorption, particularly after inspiring pure oxygen during preoxygenation and before OLV. The end-expiratory volume of the dependent lung therefore trends toward residual volume in the absence of positive pressure to counteract these factors. Despite maintaining positive end-expiratory pressure (PEEP), a pig model of OLV demonstrates that much of the ventilated lung remains suboptimally aerated at end-expiration ( Fig. 66.2 ). Position of the patient is an additional factor influencing the distribution of pulmonary blood flow, due to the effect of gravity. More recent studies call into question the magnitude of this effect on distribution of pulmonary perfusion as opposed to other characteristics, suggesting that gravity’s effects are superimposed on a background of heterogeneous perfusion due to the underlying structure of the vascular bed. At least to some extent, lateral positioning does allow for a gravity-induced increase in perfusion to the dependent ventilated lung, thereby minimizing shunt.

The factors shown can influence pulmonary blood flow distribution during OLV in the lateral position according to the direction of the adjacent arrows. HPV, Hypoxic pulmonary vasoconstriction; PVR, pulmonary vascular resistance.

From Slinger PD, Campos JH: Anesthesia for thoracic surgery. In Miller RD, editor: Miller’s anesthesia, 8th ed. Philadelphia, Elsevier, 2015, pp 1942–2006.

A–F, A pig model of OLV shows that with PEEP of 5, there remains significant alveolar collapse at end-expiration with tidal volumes of both 5 and 10 mL/kg. In the period after OLV, there is more collapse in the higher tidal volume (Vt) group, which is attributed to repeated tidal recruitment with each breath.

From Lohser J, Slinger P: Lung injury after one-lung ventilation: a review of the pathophysiologic mechanisms affecting the ventilated and the collapsed lung. Anesth Analg 121[2]:302-318, 2015.

The pulmonary vasculature itself is not a passive bystander; on the contrary, it assumes an active role in determining regional perfusion via hypoxic pulmonary vasoconstriction (HPV). In response to low alveolar oxygen tensions, pulmonary arterial smooth muscle constricts increasing regional resistance though the pulmonary vascular bed and redistributing flow away from hypoxic areas. This mechanism optimizes V/Q matching and is most effective when the hypoxic fraction of lung tissue corresponds to between 30% and 70% of the lung. The onset of this response is rapid, plateauing over approximately 20 minutes, followed by a delayed phase beginning after 40 minutes. The magnitude of this response in experimental settings (breathing 5% O ₂ into the test lung) has been shown to increase pulmonary vascular resistance in the test lung by threefold to fivefold and decrease perfusion to that lung by 30% to 45%. Debate exists in the literature with regard to the presence of a preconditioning effect whereby the sensitivity of the response to hypoxia increases on repeat exposure; these disparities in results may reflect differences in experimental conditions such as duration of hypoxic exposure and interval before reexposure. HPV is attenuated by ACE inhibitors.

Surgical manipulation itself may influence the distribution of pulmonary blood flow. Although local release of vasoactive mediators may oppose HPV, retraction and mechanical interference with pulmonary blood flow can decrease perfusion of the nondependent (operated) lung.

Risk Assessment

As mentioned, the majority of patients when maintained on an Fi o ₂ of 100% with adequate positioning of a double-lumen tube or bronchial blocker will not have significant hypoxemia. Several factors listed in Box 66.1 have been identified that may predict desaturation during OLV.

Most of these factors relate directly to the proportion of blood that will be shunting through the nonventilated lung. A higher percentage of ventilation or perfusion to the operative lung and surgery on the larger right lung result in relatively more blood being shunted through the collapsed lung. Supine positioning will remove the beneficial effect of gravity to increase perfusion to the dependent ventilated lung.

Perhaps surprisingly, obstructive airflow limitation correlates inversely with risk of desaturation on OLV. Mechanisms that have been proposed include delayed collapse of the nondependent lung, persistent end-expiratory airflow in the dependent lung contributing to auto-PEEP, and altered mechanisms of hypoxic pulmonary vasoconstriction. The most significant factor correlating with hypoxia on OLV is the intraoperative Pa o ₂ on two-lung ventilation (TLV) before OLV. The TLV Pa o ₂ serves as a gauge of the ability of the respiratory system to optimize V/Q matching when faced with factors that promote mismatch such as general anesthesia, positive-pressure ventilation, and the lateral position.

As most of these factors are known well in advance of the surgery, this allows the anesthesiologist to be prepared for potential difficulty with oxygenation during such cases and apply prophylactic measures for prevention of hypoxia as outlined in the following section.