What's New in Critical Illness and Injury Science? The ongoing debate on the optimal resuscitative fluid and monitoring parameters.

Acid base imbalances are one of the most common disorders among critically ill patients. The ability to recognize acid base abnormalities with accuracy and ease has been the focus of significant research over the past decades. At the bedside, lactate and acid base status is also used to guide resuscitation in many scenarios including in septic shock, perioperative fluid administration, and diabetic ketoacidosis.

Two distinct theories have been suggested for identification of acid base disorders. The first is based on the Henderson–Hasselbach equation and assesses the metabolic components of acid-base physiology based on serum bicarbonate (HCO3-) concentrations.[1] While this approach has existed since 1908, there are several assumptions that may limit diagnostic performance. The most significant is that this equation neglects the dissociation of the acid, the hydrolysis of the base and the dissociation of water itself. The commonly used traditional approach to acid base assessment may therefore also involve base excess (BE) and anion gap (AG) calculations to improve the merit of this approach. While this method benefits from ease of use, hypoalbuminemia in the critically ill population has been demonstrated to confound acid-base status as albumin is a weak, nonvolatile acid.

A second approach utilizes the physicochemical principles of aqueous solutions in blood.[23] The model, known as the Stewart approach, involves assessment of pCO2, the strong ion difference (SID) and the concentration of nonvolatile weak acids. The SID is the difference in strong cations (Na+, K+, Ca2+, and Mg2+) and strong anions (Cl-). While this approach has been shown to surpass the traditional approach in identification of severe disturbances, it is cumbersome and has limited application at the bedside. In addition, Dubin and colleagues found that when the traditional approach included the AGcorrected (AG corrected for albumin), the diagnostic utility of the Stewart method no longer offered any significant benefits.[4] Due to the complexity of the Stewart approach, several studies have since evaluated simplified equations to improve the bedside utility of the physiocochemical approach to acid-base assessment.[56]

While controversy still remains on the best method for detection of acid-base derangements, recognition of these disorders remains important for the assessment of perfusion and adequacy of resuscitation. BE and serum lactate have been consistently been shown to accurately predict both morbidity and mortality. In this issue of the International Journal of Critical Illness and Injury Science, Ahmed and colleagues utilize a simplified Fencl-Stewart approach previously described by Story and colleagues to compare resuscitation with Ringer's lactate and normal saline administered in 300 patients with abdominal sepsis. The authors reported that resuscitation with 20 ml/kg normal saline was associated with the development of significant nongap metabolic acidosis when compared with Ringer's lactate. These results confirmed the common teaching that administration of NS for large volume resuscitation may be associated with a nongap hypochloremic acidosis. However, it is important to note that 20 ml/kg is more comparable to smaller resuscitative volumes used for disorders that historically rely on normal saline for fluid administration, including septic shock. Therefore the question arises whether normal saline should be utilized for resuscitation at all, or if Ringer's lactate or other balanced crystalloids (i.e., Plasma-Lyte) should become the gold standard.

While the authors conclude that the simplified Fencl-Steward equation was effective and easy method to identify mixed acid-base disorders, a more crucial conclusion should focus on the differences observed in resuscitative methods (i.e., Ringer's lactate vs. normal saline). Previously published studies have evaluated the impact of various crystalloids on acid-base status in other populations including kidney transplantation and diabetic ketoacidosis.[789] These studies have consistently reported a reduction in the incidence of hyperchloremia and associated metabolic acidosis with balanced solutions (Ringer's lactate and Plasma-Lyte) over normal saline. However, all of these studies have similar limitations in that they do not report other traditional measurements of goal directed resuscitation (i.e., urine output, central venous pressure) and outcomes data is limited. Previous studies have compared crystalloid and colloid resuscitation, but high quality data evaluating clinical outcomes based on resuscitation with various crystalloids (Ringer's lactate, normal saline, Plasma-Lyte) are limited. Yunos and colleagues compared a chloride rich fluid administration (e.g., normal saline) to a period in which chloride poor fluids were preferred (i.e., Plasma-Lyte, Hartmann solution).[10] While this study failed to evaluate the impact on acid-base status, the chloride restrictive strategy was associated with a significant decrease in the incidence of acute kidney injury as defined by the RIFLE criteria. Based on the consistent reporting of acid-base derangements associated with normal saline based resuscitation, along with a higher incidence of acute kidney injury, further research to evaluating both clinical outcomes associated with normal saline vs. Ringer's lacte/Plasma-Lyte fluid resuscitation, as well as alternative resuscitation monitoring parameters (e.g. acid-base status) are warranted.

Ahmed and colleagues should be applauded for their significant contribution to the literature in the both the area of acid-base assessment and the role of balanced crystalloids for fluid resuscitation. Although it is well documented that large volumes of normal saline for fluid resuscitation may be associated with unwanted acid-base changes, there is limited data regarding the clinical significance of these abnormalities in critically ill patients. While the authors recognize that more research in these areas is warranted, their work has added significantly to the growing body of literature questioning the role of normal saline for fluid resuscitation. Based on the results reported by Ahmed and colleagues, further study on the avoidance of hyperchloremic metabolic acidosis during fluid resuscitation is warranted.