Assessing acid base disorders

Transcription

1 & 2009 International Society of Nephrology Assessing acid base disorders Horacio J. Adrogué 1,2,3, F. John Gennari 4, John H. Galla 5 and Nicolaos E. Madias 6,7 1 Department of Medicine, Baylor College of Medicine, Houston, TX, USA; 2 Department of Medicine, Methodist Hospital, Houston, TX, USA; 3 Renal Section, Veterans Affairs Medical Center, Houston, TX, USA; 4 Department of Medicine, University of Vermont College of Medicine, Burlington, VT, USA; 5 Department of Medicine, University of Cincinnati College of Medicine, Cincinnati, OH, USA; 6 Department of Medicine, Tufts University School of Medicine, Boston, MA, USA and 7 Division of Nephrology, Department of Medicine, Caritas St Elizabeth s Medical Center, Tufts University School of Medicine, Boston, MA, USA Effective management of acid base disorders depends on accurate diagnosis. Three distinct approaches are currently used in assessing acid base disorders: the physiological approach, the base-excess approach, and the physicochemical approach. There are considerable differences among the three approaches. In this, we first describe the conceptual framework of each approach, and comment on its attributes and drawbacks. We then highlight the application of each approach to patient care. We conclude with a brief synthesis and our recommendations for choosing an approach. Kidney International (2009) 76, ; doi: /ki ; published online 7 October 2009 KEYWORDS: base-excess approach; physicochemical approach; physiological approach; Stewart approach Management of acid base disorders begins with accurate diagnosis, a process requiring two tasks: First, reliable measurement of acid base variables in the blood, a complex fluid containing multiple ions and buffers; this task is an exercise in chemistry. Second, proper interpretation of the data in relation to human health and disease allowing definition of the patient s acid base status; this is an exercise in pathophysiology. The patient s history, physical examination, and additional laboratory testing and imaging, as appropriate, then help the clinician to identify the specific cause(s) of the acid base disturbance, and from that information to undertake appropriate intervention. 1 Three distinct approaches are currently used in assessing acid base disorders, each with a considerable following worldwide. For the purposes of this, we name them the physiological approach, pioneered by Van Slyke and coworkers; 2,3 the base-excess approach, developed by Astrup and co-workers; 4,5 and the physicochemical approach, proposedbystewartandextendedbyhisfollowers. 6 9 The last and newest approach has steadily gained acceptance, especially among critical-care physicians and anesthesiologists. The three approaches differ considerably. In this, we first describe the conceptual framework of each approach, and its attributes and drawbacks. We then highlight the application of each approach to patient care. We conclude with a brief synthesis and our recommendations for choosing an approach. Correspondence: Nicolaos E. Madias, Department of Medicine, Caritas St Elizabeth s Medical Center, 736 Cambridge Street, Boston, MA 02135, USA. nicolaos.madias@caritaschristi.org Received 24 June 2009; revised 14 July 2009; accepted 14 July 2009; published online 7 October 2009 PHYSIOLOGICAL APPROACH Conceptual framework The physiological approach considers acids as hydrogen ion (H þ ) donors and bases as H þ acceptors. 10 It uses solely the carbonic acid/bicarbonate buffer system for assessing acidbase status, a position rooted in the isohydric principle. Adoption of this buffer system reflects its abundance, physiological preeminence, and the fact that its two components undergo homeostatic control. 1 3 Blood ph is viewed as being determined by the prevailing levels of carbonic acid (that is, PaCO 2, the respiratory component) and plasma bicarbonate concentration ([HCO 3 ], the metabolic component, Table 1), as stipulated by the Henderson equation, [H þ ] ¼ 24 PaCO 2 /[HCO 3 ]. Kidney International (2009) 76,

2 HJ Adrogué et al.: Assessing acid base disorders Table 1 Assessment of the metabolic component of acid-base status Approach Variable Determination Remarks Physiological Plasma [HCO 3 ] Measured ph and PCO 2 Interpretation complemented by evaluation of plasma anion gap, [Na + ] ([Cl ]+[Total CO 2 ]) Base excess Blood base excess (BE) CO 2 equilibration method or calculated from measured ph and PCO 2 BE is a measure of the metabolic component of acid base status as reflected in whole blood Interpretation complemented by evaluation of plasma Physicochemical Standard BE (SBE) SID a (apparent strong ion difference) SID e (effective strong ion difference) Calculated from measured ph, PCO 2, and hemoglobin ([Na + ]+[K + ]+[Ca ++ ]+[Mg ++ ]) ([Cl ]+[lactate ]) ([Na + ]+[K + ]) ([Cl ]+[lactate ]+[other strong anions]) ([Na + ]+[K + ]) [Cl ] [HCO 3 ]+[Alb ]+[Pi ] where: [Alb ]=[Alb, g/l] [(0.123 ph) 0.631] [Pi ]=[Pi, mmol/l] [(0.309 ph) 0.469] anion gap SBE is a measure of the metabolic component of acidbase status as reflected in the extracellular compartment. It is usually calculated automatically from arterial blood gas results, but it can also be obtained using the blood acid base nomogram with the hemoglobin set at 5 g/dl 42 Interpretation complemented by evaluation of plasma anion gap These three formulas for SID a, as well as additional variants, are currently in use. SID a is mathematically equivalent to the plasma buffer base of Singer and Hastings 64 Represents the sum of plasma [HCO 3 ] and nonbicarbonate buffers (anionic equivalency of albumin and phosphate) SIG (strong ion gap) SID a SID e An estimate of the concentration of unmeasured anions in plasma that resembles the plasma anion gap Value depends upon the variant of SID a used A Tot (total concentration of weak acids in plasma) All variables and electrolytes listed are expressed in meq/l, unless otherwise indicated [total protein, g/dl] Primarily related to albumin concentration For clinical purposes, approximated by the concentration of total protein The physiological approach recognizes four acid base disorders 1,11 13 (Table 2). Metabolic disorders are expressed as primary changes in plasma [HCO 3 ], whereas respiratory disorders are expressed as primary changes in PaCO 2. Each primary change in either plasma [HCO 3 ] or PaCO 2 elicits in vivo a secondary response in the other variable that tends to minimize the change in acidity. 1,11 These secondary responses, otherwise referred to as compensatory, have been quantitated in animals and humans We discourage use of the term compensatory, because the secondary responses occasionally can yield a maladaptive effect on blood ph. 25,26 Absence of an appropriate secondary response denotes the co-existence of an additional simple acid base disorder. Use of ventilator support in critically ill patients can, of course, alter or prevent expression of the secondary changes in PaCO 2 in response to metabolic acid base disorders. These ventilator-induced alterations are viewed as complicating primary respiratory acid base disorders. The simultaneous presence of two or more simple acid base disorders defines a mixed acid base disorder. Assessment of the metabolic component is complemented by evaluating the plasma anion gap (AG), defined as [Na þ ] ([Cl ] þ [TCO 2 ]), 27 where [TCO 2 ] indicates venous total CO 2 concentration (Table 1 and Figure 1). The average normal value for plasma AG differs among health-care facilities because of methodological variation. 27 Normally, approximately 75% of the plasma AG is determined by plasma albumin concentration. 27 Thus, the plasma AG must be adjusted by subtracting or adding 2.5 meq/l from the calculated value for each 1 g/dl of plasma albumin below or above the average normal value of 4.5 g/dl, respectively. Changes in blood ph elicit small, directional changes in the anionic charge of plasma albumin and thus the AG, but these changes are ignored in clinical practice. 28,29 The anionic charge of plasma albumin decreases by only 1.5 meq/l when blood ph decreases from 7.40 to Attributes and drawbacks The physiological approach considers the acid base status of body fluids as being determined by net H þ balance (that is, influx minus efflux) and the prevailing complement of body buffers. 31,32 The chemistry of acids and bases is blended with the empirically derived secondary responses of the intact organism to primary changes in PaCO 2 or plasma [HCO 3 ]. This approach is simple regarding data collection and clinical application. The standard blood gas analyzer measures ph and PCO 2, from which plasma [HCO 3 ] is calculated (Table 1). Comparing plasma [HCO 3 ] with measured [TCO 2 ] in venous blood validates this derived variable. 1 Furthermore, most acid base disorders are first recognized by clinicians through abnormalities in venous [TCO 2 ]. Although PCO 2 is universally considered as an appropriate index of the respiratory component, plasma [HCO 3 ] has been viewed by some as an unsuitable indicator of the metabolic component. 33,34 Criticisms include lack of independence of plasma [HCO 3 ] from the respiratory component and failure of quantitation of buffers other than bicarbonate. Plasma [HCO 3 ] is certainly affected by changes 1240 Kidney International (2009) 76,

3 HJ Adrogué et al.: Assessing acid base disorders Table 2 Classification of acid base disorders Disorder Physiological approach Base-excess approach Physicochemical approach Metabolic acidosis Primary k in [HCO 3 ] k in ph; secondary k in PaCO 2 Primary base deficit ( BE, SBE) k in ph; secondary k in PaCO 2 Primary ksid e, kph Secondary DPaCO 2 /D[HCO 3 ]=k1.2 mm Hg per DPaCO 2 /DSBE=k1.0 mm Hg per Secondary DPaCO 2 not defined response meq/lk meq/lk Evaluation of plasma unmeasured anions Effect of [albumin] on acid base status Plasma AG adjusted for plasma [albumin] Normal AG acidosis (hyperchloremic High AG acidosis (normochloremic Plasma AG adjusted for plasma [albumin] Normal AG acidosis (hyperchloremic High AG acidosis (normochloremic SID acidosis where SIG=0, (ksid a =ksid e ), equivalent to hyperchloremic acidosis SIG acidosis where msig (unchanged SID a and decreased SID e ), equivalent to normochloremic acidosis No significant effect No significant effect Primary m in A tot (hyperalbuminemic Metabolic alkalosis Primary m in [HCO 3 ] m in ph; secondary m in PaCO 2 Primary base excess (+BE, +SBE) m in ph; secondary m in PaCO 2 Primary m in SID a and SID e m in ph (SID alkalosis) Secondary DPaCO 2 /D[HCO 3 ]=m0.7 mm Hg per DPaCO 2 /DSBE=m 0.6 mm Hg per Secondary DPaCO 2 not defined response meq/lm meq/lm Effect of [albumin] on acid base satus No significant effect No significant effect Primary k in A Tot (hypoalbuminemic alkalosis) Respiratory acidosis Primary m in PaCO 2 k in ph; secondary m in [HCO 3 ] Secondary D[HCO 3 ]/DPaCO 2 = response m0.1 meq/l per mm Hg m(acute) m0.3 meq/l per mm Hg m(chronic) Primary m in PaCO 2 k in ph DSBE=0 (acute) DSBE=+SBE (chronic) DSBE/DPaCO 2 =m0.4 meq/l per mm Hg m Primary m in PaCO 2 k in ph Not considered Respiratory alkalosis Primary k in PaCO 2 m in ph; secondary k in [HCO 3 ] Secondary D[HCO 3 ]/DPaCO 2 = response k0.2 meq/l per mm Hg k (acute) k0.4 meq/l per mm Hg k (chronic) Primary k in PaCO 2 m in ph DSBE=0 (acute) DSBE= SBE (chronic) DSBE/DPaCO 2 =k0.4 meq/l per mm Hg k Primary k in PaCO 2 m in ph Not considered AG, anion gap; SBE, standard base excess; SID a, apparent strong ion difference; SID e, effective strong ion difference; SIG, strong ion gap (see Table 1 for definitions). Plasma AG can be adjusted for plasma [albumin] by subtracting or adding 2.5 meq/l from the calculated value for each 1 g/dl of plasma albumin below or above the average normal value of 4.5 g/dl, respectively. Arrows pointing down (k) indicate decreases and arrows pointing up (m) indicate increases. In respiratory acid base disorders, the term acute refers to a duration of minutes to several hours. The term chronic refers to a duration of several days or longer. in PCO 2 and vice versa, because HCO 3 and H 2 CO 3 are members of a single buffer pair. 1 3 Also, sustained changes in PCO 2 substantially modify renal acidification thereby decreasing plasma [HCO 3 ] in chronic hypocapnia and increasing it in chronic hypercapnia. 14,15,25,26,32,35 Conversely, changes in plasma [HCO 3 ] modify alveolar ventilation resulting in changes in PCO 2. 16,24 However, the interdependency of PCO 2 and plasma [HCO 3 ] does not undermine the rigor of the physiological approach, because the secondary responses of the one variable to changes in the other have been quantitated for each disorder (Table 2). Regarding the second criticism, non-bicarbonate buffers are not quantitated in the physiological approach, but this does not vitiate diagnosis, because the level of plasma [HCO 3 ] always reflects the status of those buffers. Nor does it affect patient management, because the apparent bicarbonate space, used to compute acid or alkali replacement, incorporates the contribution of non-bicarbonate buffers Alterations in acid base equilibrium result in changes in the apparent bicarbonate space that have been defined experimentally and take into account the participation of non-bicarbonate buffers. 36 BASE-EXCESS APPROACH Conceptual framework The base-excess approach also advocates the centrality of H þ balance in defining acid base status. 4,5,33 In this approach, the three relevant acid base variables are blood ph, PCO 2, and base excess (BE) (Table 1). Blood BE was introduced to replace plasma [HCO 3 ] with a measure of the metabolic component that is independent from the respiratory component. Base excess represents the amount of acid or alkali that must be added to 1 l of blood exposed in vitro to a PCO 2 of 40 mm Hg to achieve the average normal ph of Acid is required when blood ph is higher than 7.40 (positive BE or base excess), whereas alkali is needed when blood ph is lower than 7.40 (negative BE or base deficit). Under normal circumstances, the average blood BE is zero. 39 During in vitro blood titration, any CO 2 -induced increase in plasma [HCO 3 ] is attended by an equivalent decrease in the anionic charge of Kidney International (2009) 76,

4 HJ Adrogué et al.: Assessing acid base disorders cations Phosphate anions cations Phosphate anions cations Phosphate anions K + 4 Proteins K + 4 K + 4 SIG AG HCO 3 24 SID a SID e Proteins HCO 3 4 AG SID a SID e Proteins HCO 3 10 AG SID a SID e Na Cl 106 Na Cl 106 Na Cl 120 Normal High AG (normochloremic) metabolic acidosis or SIG acidosis Normal AG (hyperchloremic) metabolic acidosis or SID acidosis Figure 1 Schematic illustration of plasma cations and anions in normal acid base status, high-anion-gap (normochloremic) metabolic acidosis or SIG acidosis, and normal anion gap (hyperchloremic) metabolic acidosis or SID acidosis. The numbers within the bars give ion concentrations in millimoles per liter. AG, anion gap; SID, strong ion difference; SID a, apparent strong ion difference; SID e, effective strong ion difference; SIG, strong ion gap. non-bicarbonate buffers (largely hemoglobin) that results from binding the H þ released from carbonic acid; as a result, blood BE remains constant. Opposite reactions occur during in vitro decreases in PCO 2 again resulting in a constant blood BE. In contrast, when the PCO 2 is varied in vivo as a result of hypoventilation or hyperventilation, blood BE does not remain constant because a concentration gradient for bicarbonate develops between blood and interstitial compartment; bicarbonate is lost from plasma into the interstitial fluid in hypercapnia causing a negative BE, whereas bicarbonate is added to plasma from the interstitial fluid in hypocapnia resulting in a positive BE. 19,40 This pitfall was addressed by introducing the extracellular BE or standard BE (SBE), a measure of the metabolic component that is modeled by diluting the blood sample threefold with its own plasma or estimated by using the blood BE at a hemoglobin concentration of 5 g/dl. 33,41 43 Currently, many blood gas analyzers calculate SBE from measured ph, PCO 2, and hemoglobin 44 (Table 1). The base-excess approach recognizes four acid base disturbances 11 13,45 47 (Table 2). Metabolic disorders are defined by primary changes in BE or SBE, whereas respiratory disorders represent primary changes in PCO 2. Recomputing data collected according to the physiological approach 47 allows quantitation of the secondary responses to primary changes in SBE or PCO 2 (Table 2). Assessment of SBE is complemented by evaluating the plasma AG. Attributes and drawbacks This approach champions SBE as a measure of the contribution of all extracellular buffers to a metabolic acid or alkali load that is independent of the respiratory component. Some practitioners favor this approach because SBE simplifies the estimation of acid or alkali replacement. The base-excess approach weds the chemistry of acids and bases with the anticipated secondary responses to acid base stresses. It is also relatively simple and uses the blood gas analyzer to provide all three relevant variables (Table 1). Drawbacks of this approach are the failure to include the contribution of intracellular buffers, and the presumption of a constant 1:3 ratio between blood and interstitial volume, a ratio that decreases in patients with severe edematous states. 40 Most important, the independence of SBE from the respiratory component is limited to acute hypercapnia 1242 Kidney International (2009) 76,

5 HJ Adrogué et al.: Assessing acid base disorders and hypocapnia. Chronic changes in PCO 2 elicit parallel changes in SBE by altering renal acidification. 14,15,35 Further, changes in SBE characteristic of metabolic disorders originate in part from the prevailing secondary hypocapnia or hypercapnia. 25,26 PHYSICOCHEMICAL APPROACH Conceptual framework The physicochemical approach differs fundamentally because, in addition to the influence of PCO 2 and non-volatile weak acids, it proposes that the [H þ ] of living organisms reflects changes in the dissociation of water induced by the presence of strong ions (that is, ions fully dissociated at the ph of body fluids). 6 9,48 50 Applying the principles of electroneutrality and mass conservation, Stewart examined the dissociation of water in a flask, its interaction with solutes that dissociate in solution (including the weak acids normally present in plasma), and the effects of adding CO 2. He concluded that the dissociation of water, and thus the prevailing [H þ ], is determined by the interplay of three variables defined as independent and proposed to account for all acid base effects in the solution: The strong ion difference (SID), the total concentration of weak acids (A Tot, which includes proteins and phosphate), and PCO The SID represents the sum of the concentrations of the strong cations (Na þ, K þ, Ca þþ, and Mg þþ ) minus the sum of the concentrations of the strong anions (Cl,SO ¼ 4, and anions of organic acids). By virtue of their magnitude, the concentrations of Na þ and Cl are the main determinants of SID. Stewart developed a set of six equations that, when solved simultaneously, yield the [H þ ] of the solution. 6 8 The calculated value equals measured acidity using a ph electrode. In this approach, both [H þ ] and [HCO 3 ] are defined as dependent variables, their values being determined exclusively by the three independent variables. Bicarbonate is actually viewed as an irrelevant measure of acid base status 6 that simply contributes to filling the gap between strong cations and strong anions. The physicochemical approach proposes that acid base disorders be assessed using SID and A Tot (collectively representing the metabolic component), and PCO 2 (respiratory component). Clinical application of this approach involves measurement of blood ph and PCO 2, and determination of the two formulations of SID and A Tot. 30,51,52 Subtracting strong anions from strong cations yields the apparent SID (SID a ). The simplest estimation of SID a involves subtracting [Cl ] from the sum of [Na þ ] and [K þ ], 52 and under normal conditions, it equals approximately 40 meq/l (Table 1 and Figure 1). The other formulation of SID, effective SID (SID e ), represents the sum of plasma [HCO 3 ] and the anionic equivalency of albumin and phosphate (Table 1 and Figure 1); it is estimated from blood ph, PCO 2, and the plasma concentrations of albumin and phosphate using a formula or a nomogram. 51 The difference between SID a and SID e, known as strong ion gap (SIG), is proposed as a more accurate estimate of plasma unmeasured anions than plasma AG. Under normal conditions, SIG equals zero. In organic acidosis SID a remains unchanged, but it decreases in hyperchloremic metabolic acidosis. On the other hand, SID e decreases in both types of metabolic acidosis. SIG equals zero in hyperchloremic metabolic acidosis (SID acidosis, Figure 1), 51,53 whereas it increases in organic acidosis (SIG acidosis, Figure 1) reflecting high levels of unmeasured anions, such as lactate and ketoanions. Estimation of A Tot for clinical purposes requires measurement of plasma total proteins or plasma albumin (Table 1). The physicochemical approach defines six acid base disorders according to primary deviations in each of the three independent variables 30,51 (Table 2). Abnormal levels of SID e or A Tot indicate the presence of metabolic acid base disorders. A low SID e signifies metabolic acidosis (SID and a high SID e metabolic alkalosis (SID alkalosis). A low-sid e metabolic acidosis can occur in association with a high SIG (for example, diabetic keto or a normal (that is, zero) SIG (for example, diarrhea) (Figure 1). A high A Tot signifies metabolic acidosis (hyperalbuminemic and a low A Tot metabolic alkalosis (hypoalbuminemic alkalosis) Increases in PCO 2 define respiratory acidosis and decreases in PCO 2 respiratory alkalosis. Quantitative measures of the secondary responses to primary changes in SID e,a Tot, and PCO 2 are not available. Attributes and drawbacks The physicochemical approach provides a detailed description of acid base variables. Proponents have argued that the increased complexity required for calculating SIG is justified because it yields an index of unmeasured anions after discounting the contribution of albumin and phosphate. Some data suggest that SIG better predicts the risk of death in critically ill patients than serum lactate, AG, or blood BE. 53,59 However, most studies have not identified any diagnostic or prognostic advantage of the physicochemical approach over the other approaches in such patients. 60,61 Compared with the physiological and base-excess approaches, the physicochemical approach requires additional measurements of multiple ions and the use of computer programs 33,62,63 (Table 1). Interpreting SID is cumbersome, because its two formulations, SID a and SID e, have different connotations. The SID a itself is somewhat confusing, variable electrolytes being used in defining it, such that even the normal baseline differs substantially (Table 1). Despite claims that bicarbonate is an irrelevant anion in terms of acid base assessment, it is a major component of SID e (Figure 1). Compared with plasma [HCO 3 ] and SBE, the reliability of SID a, SID e, and A Tot is precarious, because their determination requires multiple measurements and calculations incorporating several assumptions. The classification of metabolic acid base disorders is unduly complex. As examples, metabolic acidosis can be associated with normal SID a, low SID a, normal SIG, high SIG, or high A Tot. Also, the designation of A Tot acidosis and Kidney International (2009) 76,

6 HJ Adrogué et al.: Assessing acid base disorders A Tot alkalosis, based on increased and decreased levels of plasma albumin, respectively, is largely groundless. 33 Despite in vitro data showing that changes in albumin affect acidity, there is no evidence that the body and, in particular the liver, regulates albumin to maintain acid base balance. In vivo, changes in serum albumin do not correlate with changes in PCO 2 or ph. 63 Contrary to the physiological and base-excess approaches that maintain a sharp distinction between diagnosis and cause, the physicochemical approach attempts to blend these elements with respect to metabolic disturbances, a potential source of confusion. The physicochemical approach is computationally precise but anchored exclusively in chemistry. The mathematical precision of this approach does not validate the proposed cause and effect relationship with acid base variables. To posit that SID and A Tot determine mechanistically plasma [HCO 3 ] and ph at the physiological level is presumptuous. Experimental support for such a cause and effect relationship is lacking. 63 Rather than being causes of acid base perturbations, changes in SID could be mere reflections of these perturbations. Proponents of the physicochemical approach argue that rapid infusion of isotonic saline (for example, 30 ml/kg/h), a solution with a SID of zero, generates metabolic acidosis because of a hyperchloremia-induced decrease in SID. This phenomenon is essentially prevented during rapid infusion of Ringer s lactate, a solution with an SID of 23 meq/l. 65 The physiological approach offers, however, an equally straightforward explanation. The metabolic acidosis is caused by the sizable expansion of the extracellular fluid with a solution devoid of bicarbonate (that is, isotonic saline) that results in a decrease in plasma [HCO 3 ] without a proportional decrease in PCO 2 (dilution. 66 Such a dilution of the extracellular bicarbonate stores is obviated during rapid infusion of Ringer s lactate, because the infused lactate represents a bicarbonate precursor, its metabolism generating equivalent molecules of bicarbonate. Dilution acidosis is, of course, a short-lived phenomenon in patients with normal renal function, because regulatory changes intervene rapidly; it can only be sustained in patients with advanced renal failure. CLINICAL APPLICATIONS Let us now evaluate the acid base status of two clinical cases using each of the three approaches. The laboratory data for each case and the corresponding acid base diagnoses according to each approach are shown in Table 3. Case 1 A 27-year-old woman with type 1 diabetes mellitus was brought to the emergency department with obtundation, severe hyperpnea, and blood pressure of 90/45 mm Hg. Using the physiological approach, the plasma [HCO 3 ]of 4 meq/l and blood ph of 7.05 signify metabolic acidosis. For a D[HCO 3 ] of 20 meq/l (24-4), the expected DPaCO 2 is 24 mm Hg (1.2 20, Table 2) predicting a PaCO 2 of 16 mm Hg (40-24); the patient s PaCO 2 of 15 mm Hg represents an appropriate ventilatory response. Thus, a single acid base disorder is present. The AG is markedly increased (28 meq/l) compared with the expected baseline (approximately 4 meq/l when corrected for plasma albumin concentration) indicating the presence of increased unmeasured anions in plasma. The estimated change in AG (DAG) equals 24 meq/l and approximates the decrease in plasma [HCO 3 ]. Using the base-excess approach, the SBE of 25 meq/l in conjunction with a blood ph of 7.05 is indicative of metabolic acidosis. The DPaCO 2 of 25 mm Hg is appropriate for the DSBE of 25 meq/l (expected value, 1 25, Table 2) and represents secondary hypocapnia. Thus, a single acidbase disorder is considered present. The DAG of 24 meq/l (evaluated in a manner identical to the physiological approach) points to high-anion-gap metabolic acidosis. Using the physicochemical approach, the decreased SID e of 11 meq/l is diagnostic of metabolic acidosis. The elevated SIG of 27 meq/l indicates that the acidosis is due to accumulation of unmeasured strong anions. Reflecting the low plasma albumin concentration, the decreased A Tot signifies coexisting weak-acid alkalosis (hypoalbuminemic alkalosis). The decreased PaCO 2 points to the presence of respiratory alkalosis. No information is available to judge the appropriateness of the ventilatory response. Thus, the patient has three acid base disorders. Both the physiological and the base-excess approaches diagnose an identical simple acid base disorder, namely highanion-gap metabolic acidosis. The patient s clinical history coupled with the prevailing hyperglycemia and positive tests for blood and urine ketones (data not shown) establish the presence of diabetic ketoacidosis. The focus of care is to reverse the organic acidosis. In contrast, the physicochemical approach identifies three separate acid base disorders, metabolic acidosis, hypoalbuminemic alkalosis, and respiratory alkalosis. These diagnoses can be a source of confusion for the clinician. Is the hypoalbuminemic alkalosis, a condition of questionable existence, protective in this case? Must the respiratory alkalosis be treated or is the hypocapnia appropriate for the metabolic acidosis? Attempts to address these questions risk mismanagement and divert attention from the key problem, reversing the organic acidosis. Case 2 A 58-year-old man with advanced chronic obstructive pulmonary disease and congestive heart failure was admitted because of anorexia and failure to thrive. His medications include bronchodilators and diuretics. The physiological approach identifies marked elevations in plasma [HCO 3 ] and PaCO 2 in the company of a near normal blood ph signifying a mixed acid base disorder. The PaCO 2 of 58 mm Hg exceeds the anticipated ventilatory response to metabolic alkalosis; for a D[HCO 3 ] of 11 meq/l (35-24), the expected DPaCO 2 is approximately 8 mm Hg (0.7 11, Table 2) predicting a PaCO 2 of 48 mm Hg (40 þ 8). Thus, both metabolic alkalosis and respiratory acidosis are present. The increased plasma [HCO 3 ] represents the sum of the effects of diuretics and chronic hypercapnia on renal 1244 Kidney International (2009) 76,

7 HJ Adrogué et al.: Assessing acid base disorders Table 3 Clinical examples Mean normal values Case 1 Case 2 Measured variables ph PaCO 2 40 mm Hg [HCO 3 ] 24 meq/l [Na + ] 140 meq/l [K + ] 4.0 meq/l [Cl ] 104 meq/l [TCO 2 ] 26 meq/l [Albumin] 4.5 g/dl Pi 1.2 mmol/l Derived variables AG 10 meq/l 28 2 SBE 0 meq/l SID a 40 meq/l SID e 40 meq/l SIG 0 meq/l 27 0 A tot 15 meq/l 7 5 Acid-base diagnoses Physiological approach One disorder: Two disorders: Metabolic acidosis (high AG metabolic Metabolic alkalosis Respiratory acidosis Base-excess approach One disorder: Two disorders: Metabolic acidosis (high AG metabolic Metabolic alkalosis Respiratory acidosis Physicochemical approach Three disorders: Two disorders: Metabolic acidosis (SIG Hypoalbuminemic alkalosis Respiratory alkalosis Hypoalbuminemic alkalosis Respiratory acidosis acidification. The AG is reduced, the result of severe hypoalbuminemia. Using the base-excess approach, one finds an SBE of 11 meq/l along with the marginally increased blood ph of 7.41, which indicates the presence of metabolic alkalosis (diuretic-induced). The DPaCO 2 of 18 mm Hg (58-40) exceeds the anticipated secondary response to the DSBE of 11 meq/l ( B7, predicting a PaCO 2 of 47 mm Hg) and signifies coexisting respiratory acidosis. The AG is reduced because of severe hypoalbuminemia. The physicochemical approach indicates that the normalcy of SID a, SID e, and SIG contrasts with a very diminished A Tot of only 5 meq/l. The patient s diagnosis is weak-acid alkalosis (hypoalbuminemic alkalosis). The increased PaCO 2 signifies respiratory acidosis. Two acid base disorders are diagnosed. Both the physiological and base-excess approaches will lead clinicians to introduce measures to improve alveolar ventilation as well as reduce the metabolic component and thus blood ph by administering potassium chloride and possibly acetazolamide. These two measures complement each other because even relative alkalemia reduces the ventilatory drive. By contrast, the physicochemical approach will fail to pursue these therapeutic measures because it cannot judge whether the increase in PaCO 2 represents primary hypercapnia, secondary hypercapnia, or both. Instead, it might prompt clinicians to raise plasma albumin level, a scientifically unsupported measure. No research has been directed toward the potential clinical implications of this divergence in diagnosis. We can only speculate whether such variable assessment of the acid base status and resultant differences in clinical management contribute to patients morbidity and mortality. SYNTHESIS AND RECOMMENDATIONS The physiological and base-excess approaches adequately fulfill both the chemical and the pathophysiological tasks to assessing acid base status in a relatively simple and straightforward manner. Conversely, the physicochemical approach addresses these two tasks by introducing considerable complexity, as it requires multiple determinations to compute its panel of acid base variables. In our judgment, this approach builds a mathematical superstructure that is superfluous, impractical, and at times misleading. The physiological approach offers quantitative measures of the secondary responses to primary acid base perturbations that have been derived from observational and experimental studies in humans (Table 2). A corresponding quantitation is provided by the base-excess approach that is based on recomputing data collected according to the physiological approach (Table 2). Conversely, no quantitative assessment of the secondary responses to primary changes in Kidney International (2009) 76,

8 HJ Adrogué et al.: Assessing acid base disorders SID e, A Tot, and PCO 2 is offered by the physicochemical approach. This situation is a major drawback and risks misdiagnosis of the secondary responses as independent, simple acid base disorders. Despite its general adequacy and simplicity, the baseexcess approach uses blood acid base nomograms obtained in vitro and assumptions that detract from its overall reliability. Only the physiological approach precisely quantitates the two components of the dominant buffer pair in blood, which are in equilibrium with the remaining body buffers, thereby providing direct and reliable assessment of acid base status in vivo. We conclude that the physiological approach remains the simplest, most rigorous, and most serviceable approach to assessing acid base disorders. DISCLOSURE All the authors declared no competing interests. REFERENCES 1. Adrogué HJ, Madias NE. Tools for clinical assessment. In: Gennari FJ, Adrogué HJ, Galla JH, Madias NE (eds). Acid-Base Disorders and their Treatment. Taylor & Francis: Boca Raton, 2005, pp Henderson LJ. The theory of neutrality regulation in the animal organism. Am J Physiol 1907; 18: Van Slyke DD, Wu H, McLean FC. Studies of gas and electrolyte equilibria in the blood. V. Factors controlling the electrolyte and water distribution in the blood. J Biol Chem 1923; 56: Astrup P. A simple electrometric technique for the determination of carbon dioxide tension in blood and plasma, total content of carbon dioxide in plasma, and bicar-bonate content in separated plasma at a fixed carbon dioxide tension (40 mmhg). Scand J Clin Lab Invest 1956; 8: Siggaard-Andersen O, Engel K, Jorgensen K et al. A micro method for determination of ph, carbon dioxide tension, base excess and standard bicarbonate in capillary blood. Scand J Clin Invest 1960; 12: Stewart PA. Independent and dependent variables of acid-base control. Respir Physiol 1978; 33: Stewart PA. How to understand acid-base. In: Stewart PA (ed). A Quantitative Acid-Base Primer for Biology and Medicine. Elsevier: New York, 1981, pp Stewart PA. Modern quantitative acid-base chemistry. Can J Physiol Pharmacol 1983; 61: Fencl V, Leith DE. Stewart s quantitative acid-base chemistry: applications in biology and medicine. Respir Physiol 1993; 91: Lowry TM. The uniqueness of hydrogen. J Soc Chem Indust 1923; 42: Andersen OS, Astrup P, Bates RG et al. Statement on acid-base terminology. Report of the ad hoc Committee of the New York Academy of Sciences Conference, November 23 24, Ann Intern Med 1965; 63: Winters RW. Terminology of acid-base disorders. Ann Intern Med 1965; 63: Elkinton JR. Acid-base disorders and the Clinician (Editorial). Ann Intern Med 1965; 63: Schwartz WB, Brackett Jr NC, Cohen JJ. The response of extracellular hydrogen ion concentration to graded degrees of chronic hypercapnia: the physiologic limits of the defense of ph. J Clin Invest 1965; 44: Gennari FJ, Goldstein MB, Schwartz WB. The nature of the renal adaptation to chronic hypocapnia. J Clin Invest 1972; 51: Madias NE, Bossert WH, Adrogué HJ. Ventilatory response to chronic metabolic acidosis and alkalosis in the dog. J Appl Physiol 1984; 56: Adrogué HJ, Madias NE. Influence of chronic respiratory acid-base disorders on acute CO 2 titration curve. J Appl Physiol 1985; 58: Madias NE, Adrogué HJ. Influence of chronic metabolic acid-base disorders on the acute CO 2 titration curve. J Appl Physiol 1983; 55: Brackett Jr NC, Cohen JJ, Schwartz WB. Carbon dioxide titration curve of normal man. Effect of increasing degrees of acute hypercapnia on acid-base equilibrium. N Engl J Med 1965; 272: Galla JH. Chloride-depletion alkalosis. In: Gennari FJ, Adrogué HJ, Galla JH, Madias NE (eds). Acid-Base Disorders and their Treatment. Taylor & Francis: Boca Raton, 2005, pp Brackett Jr NC, Wingo CF, Muren O et al. Acid-base response to chronic hypercapnia in man. N Engl J Med 1969; 280: Arbus GS, Hebert LA, Levesque PR et al. Characterization and clinical application of the significance band for acute respiratory alkalosis. N Engl J Med 1969; 280: Krapf R, Beeler I, Hertner D et al. Chronic respiratory alkalosis. The effect of sustained hyperventilation on renal regulation of acid-base equilibrium. N Engl J Med 1991; 324: Bushinsky DA, Coe FL, Katzenberg C et al. Arterial PCO 2 in chronic metabolic acidosis. Kidney Int 1982; 22: Madias NE, Schwartz WB, Cohen JJ. The maladaptive renal response to secondary hypocapnia during chronic HCl acidosis in the dog. J Clin Invest 1977; 60: Madias NE, Adrogué HJ, Cohen JJ. Maladaptive renal response to secondary hypercapnia in chronic metabolic alkalosis. Am J Physiol 1980; 238: F283 F Kraut JA, Madias NE. Serum anion gap: its uses and limitations in clinical medicine. Clin J Am Soc Nephrol 2007; 2: Adrogué HJ, Brensilver J, Madias NE. Changes in the plasma anion gap during chronic metabolic acid-base disturbances. Am J Physiol 1978; 235: F291 F Madias NE, Ayus JC, Adrogué HJ. Increased anion gap in metabolic alkalosis: the role of plasma protein equivalency. N Engl J Med 1979; 300: Fencl V, Jabor A, Kazda A et al. Diagnosis of metabolic acid-base disturbances in critically ill patients. Am J Respir Crit Care Med 2000; 162: Adrogué HJ, Madias NE. Measurement of acid-base status. In: Gennari FJ, Adrogué HJ, Galla JH, Madias NE (eds). Acid-Base Disorders and their Treatment. Taylor & Francis: Boca Raton, 2005, pp Gennari FJ. Regulation of acid-base balance: overview. In: Gennari FJ, Adrogué HJ, Galla JH, Madias NE (eds). Acid-Base Disorders and their Treatment. Taylor & Francis: Boca Raton, 2005, pp Siggaard-Andersen O, Fogh-Andersen N. Base excess or buffer base (strong ion difference) as measure of a non-respiratory acid-base disturbance. Acta Anaesthesiol Scand 1995; 39(Suppl 107): Severinghaus JW. Case for standard-base excess as the measure of nonrespiratory acid-base imbalance. J Clin Monit 1991; 7: Madias NE, Adrogué HJ, Horowitz GL et al. A redefinition of normal acid-base equilibrium in man: Carbon dioxide tension as a key determinant of plasma bicarbonate concentration. Kidney Int 1979; 16: Adrogué HJ, Brensilver J, Cohen JJ et al. Influence of steady-state alterations in acid-base equilibrium on the fate of administered bicarbonate in the dog. J Clin Invest 1983; 71: Adrogué HJ, Madias NE. Management of life-threatening acid-base disorders. (First of two parts). N Engl J Med 1998; 338: Adrogué HJ, Madias NE. Management of life-threatening acid-base disorders. (Second of two parts). N Engl J Med 1998; 338: Siggaard-Andersen O. Normal values and extreme values. In: The Acid- Base Status of the Blood, 2nd edn. William and Wilkins: Baltimore, 1964, pp Schwartz WB, Relman AS. A critique of the parameters used in the evaluation of acid-base disorders whole blood buffer base and standard bicarbonate compared with blood ph and plasma bicarbonate concentration. N Engl J Med 1963; 268: Siggaard-Andersen O. An acid-base chart for arterial blood with normal and pathophysiological reference areas. Scand J Clin Lab Invest 1971; 27: Severinghaus JW. Acid-base balance nomogram A Boston-Copenhagen détente. Anesthesiology 1976; 45: Siggaard-Andersen O. Blood acid-base alignment nomogram: scales for ph, P CO 2, base excess of whole blood of different hemoglobin concentrations, plasma bicarbonate, and plasma total-co 2. Scand J Clin Lab Invest 1963; 15: Siggaard-Andersen O. The Van Slyke equation. Scand J Clin Lab Invest 1977; 37(Suppl 146): Siggaard-Andersen O. Terminology. In: The Acid-Base Status of the Blood, 2nd edn. William and Wilkins: Baltimore, 1964, pp Siggaard-Andersen O. The acid-base status of the whole organism. In: The Acid-Base Status of the Blood, 2nd edn. William and Wilkins: Baltimore, 1964: Kidney International (2009) 76,

9 HJ Adrogué et al.: Assessing acid base disorders 47. Schlichtig R, Grogono AW, Severinghaus JW. Human PaCO 2 and standard base excess compensation for acid-base imbalance. Crit Care Med 1998; 26: Kellum JA. Metabolic acidosis in the critically ill: lessons from physical chemistry. Kidney Int 1998; 53(Suppl 66): Kellum JA. Determinants of blood ph in health and disease. Crit Care 2000; 4: Kellum JA. Clinical : reunification of acid-base physiology. Crit Care 2005; 9: Corey HE. Stewart and beyond: new models of acid-base balance. Kidney Int 2003; 64: Lloyd P. Strong ion calculator a practical bedside application of modern quantitative acid-base physiology. Crit Care Resuscit 2004; 6: Kaplan LJ, Kellum JA. Initial ph, base deficit, lactate, anion gap, strong ion difference, and strong ion gap predict outcome from major vascular injury. Crit Care Med 2004; 32: Rossing TH, Maffeo N, Fencl V. Acid-base effects of altering plasma protein concentration in human blood in vitro. J Appl Physiol 1986; 61: McAuliffe JJ, Lind LJ, Leith DE, et al. Hypoproteinemic alkalosis. Am J Med 1986; 81: Fencl V, Rossing TH. Acid-base disorders in critical care medicine. Annu Rev Med 1989; 40: Figge J, Rossing TH, Fencl V. The role of serum proteins in acid-base equilibria. J Lab Clin Med 1991; 117: Figge J, Mydosh T, Fencl V. Serum proteins and acid-base equilibria: a follow-up. J Lab Clin Med 1992; 120: Balasubramanyan N, Havens PL, Hoffman GM. Unmeasured anions identified by the Fencl-Stewart method predict mortality better than excess, anion gap, and lactate in patients in the pediatric intensive care unit. Crit Care Med 1999; 27: Dubin A, Menises MM, Masevicius FD et al. Comparison of three different methods of evaluation of metabolic acid-base disorders. Crit Care Med 2007; 35: Cusack RJ, Rhodes A, Lochhead P et al. The strong ion gap does not have prognostic value in critically ill patients in a mixed medical/surgical adult ICU. Intensive Care Med 2002; 28: Emmett M. Clinical acid-base disorders: traditional versus new analytical models. Kidney Int 2004; 65: Kurtz I, Kraut J, Ornekian V et al. Acid-base analysis: a critique of the Stewart and bicarbonate-centered approaches. Am J Physiol Renal Physiol 2008; 294: F1009 F Singer RB, Hastings AB. An improved clinical method for the estimation of disturbances of the acid-base balance of human blood. Medicine 1948; 27: Scheingraber S, Rehm M, Sehmisch C et al. Rapid saline infusion produces hyperchloremic acidosis in patients undergoing gynecologic surgery. Anesthesiology 1999; 90: Jaber BL, Madias NE. Marked dilutional acidosis complicating management of right ventricular myocardial infarction. Am J Kidney Dis 1997; 30: Kidney International (2009) 76,