Assessment of Acid-Base Status

Introduction

Maintenance of normal blood pH is critical for cellular function. Yet, blood pH is uncommonly measured in the outpatient setting. Instead, serum [total CO₂] ([TCO₂]) is used to screen for acid-base disturbances, the argument being that both metabolic and respiratory disorders are generally associated with abnormal serum [TCO₂]. Findings from recent studies in people with CKD and those on hemodialysis suggest that measurement of blood pH adds important prognostic and therapeutic information above and beyond [TCO₂]. Furthermore, urinary concentrations of ammonium and citrate are other measures of kidney acid-base regulation that have prognostic and therapeutic value in CKD. We contend that the evaluation of acid-base status should move beyond simply measuring [TCO₂]. Here, we summarize recent findings supporting comprehensive evaluations of acid-base balance in clinical practice.

Blood pH in CKD

The serum [TCO₂] measured in clinical laboratories is used as a surrogate for [HCO₃ ⁻]. A low serum [TCO₂] commonly observed in CKD is commonly viewed as indicative of metabolic acidemia, and it is associated with CKD progression. However, the potential effect of alterations of blood pH on CKD progression independent of [TCO₂] is unclear. Recently, a retrospective study of 1058 Japanese patients with CKD addressed this important question (1). In contrast to the United States, clinical laboratories in Japan measure pH and partial pressure of carbon dioxide (pCO₂) in venous blood and calculate blood [HCO₃ ⁻] using the Henderson–Hasselbalch equation. When compared with those with high serum [HCO₃ ⁻] (≥26.6 mEq/L), those with low serum [HCO₃ ⁻] (≤21.5 mEq/L) did not have a higher risk of kidney failure. Further analyses revealed, however, that those with low serum [HCO₃ ⁻] who were acidemic (pH<7.32) had a 2.29-fold higher risk of kidney failure than those with low serum [HCO₃ ⁻] but who did not have acidemia (pH≥7.32). In contrast, among patients without acidemia, no significant association was found between serum [HCO₃ ⁻] and kidney failure. These data indicate that blood pH might be an important determinant of CKD progression among those with low blood [HCO₃ ⁻], although this requires confirmation in other studies. The value of measuring blood pCO₂ is also intuitively obvious, as it allows the clinician to recognize respiratory acid-base disturbances. For example, only 59% of patients in that study with a low blood [HCO₃ ⁻] were acidemic, whereas the remaining 41% had a normal (38%) or an alkalemic blood pH (3%). Thus, in a number of patients with CKD, a low blood [HCO₃ ⁻] might be due to a composite of metabolic acidosis and respiratory alkalosis or respiratory alkalosis alone rather than pure metabolic acidosis.

These findings might have important therapeutic implications in CKD. Clinical practice guidelines suggest treating low serum [TCO₂] with base in patients with CKD. Because approximately 15% of the 37 million people with CKD in the United States have a low serum [TCO₂], this would apply to 5.5 million individuals (2). However, if the risk of CKD progression is only faster among individuals with metabolic acidemia and if the prevalence rates of acid-base disorders found were similar to that observed in Japan, then 3.3 million (9%) patients with CKD would have low serum [TCO₂] with low blood pH (metabolic acidemia), and 2.2 million (6%) would have low serum [TCO₂] with normal or high pH. This latter group of 2.2 million individuals might not benefit from base therapy, and perhaps, acidemia should be confirmed in those with low serum [TCO₂] before recommending base.

Measurement of Blood pH in Hemodialysis

Predialysis [TCO₂] is also used to evaluate acid-base status in patients on hemodialysis with the assumption that a low serum [TCO₂] represents metabolic acidosis and an elevated [TCO₂] represents metabolic alkalosis. Yet, comorbidities, such as heart failure and chronic obstructive pulmonary disease, are highly prevalent in patients on dialysis and may cause primary respiratory acid-base disturbances. Indeed, a study of 53 patients on hemodialysis from a single dialysis unit (3) revealed that respiratory acid-base disorders alone or as part of a complex acid-base disorder were found in 41% of patients, indicating that alterations in serum [TCO₂] were not solely due to a metabolic disorder. In another study, 71 samples obtained from 25 patients on maintenance hemodialysis (4) revealed that the predialysis serum [TCO₂] was not predictive of the underlying acid-base disorder in 40% of individuals. The majority of these, 36%, were misclassified as having acidemia on the basis of a low serum [TCO₂] when in fact they had normal or high pH. For those samples in which low predialysis [TCO₂] was not accompanied by acidemia, mean postdialysis pH was 7.51. Thus, increasing dialysate [HCO₃⁻] or administering supplemental base solely on the basis of the presence of a low [TCO₂] might cause excessive alkalemia. Lung disease, nutritional parameters, or residual kidney function did not explain misclassification observed in that study, indicating that these parameters cannot predict blood pH. Thus, measuring blood pH and pCO₂ is required to better inform bicarbonate therapy in patients on hemodialysis.

Importance of Blood pH in the General Population

In a study of >2000 generally healthy adults 70–79 years of age, a low [HCO₃ ⁻] (measured using arterialized venous blood gases) was associated with higher mortality, a result that was independent of blood pH (5). Further analyses revealed that mortality was higher in those with both metabolic acidosis (25% higher) and respiratory alkalosis (29% higher), compared with those with normal acid-base status. These findings demonstrate the importance of defining the risk of various acid-base disorders in healthy older individuals and the value of obtaining full acid-base data in guiding clinical decision making. The measurements of blood pH and pCO₂ could also be of value in patients with pulmonary disease, sleep apnea, or osteoporosis (6) or those who are receiving drugs (for example, diuretics or opioids) that can affect acid-base balance.

Barriers to Measuring Blood Gases in the Outpatient Setting

Although sampling arterial blood is most accurate in assessing acid-base disorders, it has several potential complications, including bleeding and pain. Peripheral venous blood gases have been shown to correlate with arterial blood gases and, thus, seem to be a reasonable substitute (7). In patients on hemodialysis, there may be different values for blood obtained from a fistula or graft than a catheter whose tip is in the superior vena cava or a femoral vein. At present, measurements are often not done in the main clinical laboratory but in the blood gas laboratory. Determination of the acid-base parameters has to be accomplished relatively soon after the blood gases are obtained to minimize errors; the provision of point-of-care devices in dialysis units or clinical laboratories could mitigate these requirements. In terms of cost, the Medicare rate for a blood gas without oxygen saturation is $26. In comparison, the rate for parathyroid hormone is $41, and the rate for 25-hydroxyvitamin D is $30. Further studies are needed to confirm the utility and practicality of various means of measuring acid-base parameters in the outpatient setting.

Other Measures of Acid-Base Balance

The term eubicarbonatemic metabolic acidosis has been used to describe a state in which systemic pH, pCO₂, and [HCO₃ ⁻] are normal, but there is acid excess in interstitial tissues that adversely affects cellular function (8). Identifying this state clinically is challenging, but two measurements, urinary ammonium and urinary citrate, are under investigation. Urinary ammonium excretion is impaired in patients with CKD, and reductions in ammonium excretion are observed before low [TCO₂] is observed (9). Furthermore, low urinary ammonium excretion is associated with higher risk of CKD progression. Urinary citrate levels are also reduced in the setting of acid retention, due to reclamation of filtered citrate. Although associations between urinary citrate levels and CKD progression or risk of developing low [TCO₂] in CKD have not been established, urinary citrate levels increase in response to base therapy among patients with clinically normal acid-base balance (10), suggesting that renal acid retention was present. Also, recent studies show that urinary citrate-creatinine ratio is a potential marker of acid retention in CKD (11). Whether it would supplant blood pH or provide an ancillary marker of the severity of body acidity remains to be determined.

We contend that measurement of blood pH and pCO₂ is necessary to precisely characterize the acid-base state of patients with CKD both with and without severe kidney failure requiring dialysis. It might also be worthwhile in individuals with conditions commonly associated with acid-base derangements, such as liver disease, or in those in whom an acidic environment might contribute to the pathogenesis of the abnormality, such as osteoporosis. Given the recognition of the potential adverse effects of eubicarbonatmeic metabolic acidosis, the utility of nontraditional measurements, such as urinary ammonium and citrate, is under active investigation. Comprehensive acid-base assessments in persons with CKD or other selected disorders for the purpose of improving clinical outcomes and improving clinical decision making with respect to administration of base therapy should be given serious consideration. The acid-base status of patients is a critical factor affecting their well-being, and continued investigation into appropriate monitoring of patients is warranted.

Disclosures

J.A. Kraut reports employment with the Veterans Administration Greater Los Angeles Healthcare System; has consulted for Tricida, Inc. (South San Francisco, CA); has received honoraria from the American Society of Nephrology, the Henry Ford Hospital, and Tricida; and reports a provisional patent submitted by the University of California, Los Angeles (Base in treatment of metabolic acidosis). K.L. Raphael reports employment with the Veterans Affairs Portland Health Care System and Oregon Health & Science University and serves as a consultant for AstraZeneca.

Funding

J.A. Kraut’s work is supported in part by unrestricted funds from the University of California, Los Angeles.