Serum potassium elevation and acute respiratory acidosis during thoracoscopic esophagectomy with intrathoracic carbon dioxide insufflation: a multicenter retrospective observational study

Technical Summary

Problem Statement
Thoracoscopic esophagectomy with intrathoracic carbon dioxide (CO₂) insufflation and lung collapse, often performed in the prone position, significantly alters respiratory physiology and acid-base balance. While acute respiratory acidosis is a known consequence of this procedure, the conventional physiological view holds that respiratory acidosis causes minimal to no increase in serum potassium compared to metabolic acidosis. However, clinical observations during these specific surgeries frequently reveal significant serum potassium elevations concurrent with acute respiratory acidosis. The mechanisms driving this phenomenon in the context of intrathoracic CO₂ insufflation and one-lung ventilation remain unclear, and the magnitude of this electrolyte shift has not been systematically quantified.

Methodology
This was a multicenter, retrospective, observational study conducted at Aichi Medical University Hospital and Aichi Cancer Center between January 2022 and December 2024. The study included 131 adult patients undergoing thoracoscopic esophagectomy with intrathoracic CO₂ insufflation in either the prone or lateral decubitus position.

• Data Collection: Arterial blood gas (ABG) variables and serum potassium were measured at two specific timepoints:
  • T1 (Baseline): After anesthesia induction, before surgical positioning or incision.
  • T2 (Intervention): During the thoracic phase under CO₂ insufflation, specifically at the timepoint of maximal respiratory challenge (defined as the sample with the lowest arterial pH).
• Primary Outcome: The paired difference in serum potassium concentration (ΔK) between T1 and T2.
• Statistical Analysis: The study employed paired t-tests for primary outcomes. Subgroup analyses were conducted based on surgical position and institution. A sensitivity analysis excluded patients with marked metabolic acidosis (base excess < −5 mmol/L). Multivariable linear regression models were constructed to identify independent factors associated with ΔK, adjusting for changes in PaCO₂, lactate, and preoperative renal function (eGFR).

Key Results

• Potassium Elevation: Serum potassium increased significantly from a baseline of 3.96 ± 0.38 mEq/L at T1 to 4.59 ± 0.63 mEq/L at T2, representing a mean increase of 0.64 mEq/L (95% CI, 0.55–0.73; p < 0.001).
• Acid-Base Status: Concurrent with the potassium rise, patients developed acute respiratory acidosis. Mean pH decreased from 7.387 to 7.247, and PaCO₂ increased from 41.48 to 58.60 mmHg.
• Metabolic Contribution: While lactate changes were independently associated with potassium changes in multivariable analysis, the potassium increase remained significant and of similar magnitude (0.618 mEq/L) even after excluding patients with marked metabolic acidosis.
• Predictors: In the multivariable model, the change in lactate (ΔLac) was the only independent predictor of ΔK (β = 0.303; p = 0.004). Neither the change in PaCO₂ (ΔPaCO₂) nor preoperative eGFR was independently associated with the potassium increase.
• Intercept Analysis: The centered multivariable model yielded a significantly positive intercept (β = 0.637), suggesting that a substantial potassium increase occurs even in patients with average values for metabolic stress, renal function, and CO₂ changes.
• Subgroups: The potassium increase was consistent across both institutions and was also observed in the smaller subgroup of patients undergoing surgery in the lateral decubitus position.

Significance and Claims
The paper claims that a clinically meaningful increase in serum potassium occurs during thoracoscopic esophagectomy with intrathoracic CO₂ insufflation, occurring simultaneously with acute respiratory acidosis. The authors argue that this elevation cannot be explained solely by metabolic stress or the degree of hypercapnia alone, as indicated by the robust increase even after excluding severe metabolic acidosis and the lack of independent association with ΔPaCO₂ in the regression model.

The study suggests that the specific intraoperative environment—characterized by hyperacute exogenous CO₂ loading, lung collapse, and positional changes—may trigger physiological mechanisms (such as immediate transcellular shifts involving intracellular H⁺ buffering and K⁺ movement) that differ from the slower compensatory mechanisms seen in typical respiratory acidosis.

Clinical Implications
The authors conclude that arterial blood gas assessment, including potassium measurement, is warranted during periods of significant hypercapnia in these procedures. They caution that relying solely on end-tidal CO₂ (EtCO₂) may underestimate the severity of acidosis due to a widened PaCO₂–EtCO₂ gradient. Furthermore, the findings suggest that potential electrolyte consequences should be considered when permissive hypercapnia is accepted, as acute potassium elevation may amplify the risk of intraoperative arrhythmias, particularly given the proximity of thoracic manipulation to the heart. The study positions this surgical setting as a unique clinical model for examining physiological responses to acute CO₂ loading that challenges conventional assumptions about respiratory acidosis and potassium homeostasis.