Original Article
Chronic Respiratory Alkalosis — The Effect of Sustained Hyperventilation on Renal Regulation of Acid–Base Equilibrium
N Engl J Med 1991; 324:1394-1401May 16, 1991
- Abstract
- Abstract
Background.
In normal subjects, chronic hyperventilation lowers plasma bicarbonate concentration, primarily by inhibiting the urinary excretion of net acid. The quantitative relation between reduced arterial carbon dioxide tension (PaCO2) and the plasma bicarbonate concentration in the chronic steady state has not been studied in humans, however, and the laboratory criteria for the diagnosis of chronic respiratory alkalosis therefore remain undefined. We wished to provide such reference data for clinical use. Moreover, because chronic hyperventilation paradoxically lowers blood pH still further in dogs with metabolic acidosis, we desired to study the effect of chronic hypocapnia on the plasma bicarbonate concentration (and blood pH) in normal human subjects in whom acidosis had been induced with ammonium chloride.
Methods.
Under metabolic-balance conditions, we used altitude-induced hypobaric hypoxia to produce chronic hypocapnia in nine normal young men, five of whom received ammonium chloride daily to cause metabolic acidosis (the mean [±SE] steady-state plasma bicarbonate level in these five was 12.0±0.5 mmol per liter).
Results.
For each decrease of 1 mm Hg (0.13 kPa) in the PaCO2, the plasma bicarbonate concentration decreased by 0.41 mmol per liter in the subjects who started with a normal plasma bicarbonate concentration and by 0.42 mmol per liter in the subjects with acidosis. In contrast to the findings in previous studies of dogs, hypocapnia increased blood pH similarly in both groups; the blood hydrogen ion concentration decreased by about 0.4 nmol per liter for every decrease of 1 mm Hg (0.13 kPa) in PaCO2.
Conclusions.
These results provide reference data for the diagnosis of chronic respiratory alkalosis in humans. Although chronic hypocapnia decreased plasma bicarbonate levels similarly in normal subjects with acidosis and without acidosis, the percent reduction in PaCO2 was always greater than the corresponding percent reduction in the plasma bicarbonate concentration. Therefore, as was not true of the response in dogs, the subjects' blood pH always increased with hyperventilation, regardless of the initial plasma bicarbonate concentration. (N Engl J Med 1991;324:1394–401.)
Article Activity
- Article
CHRONIC respiratory alkalosis is a common acid–base disturbance characterized by a primary and sustained decrease in arterial carbon dioxide tension (PaCO2) — that is, by primary hypocapnia. A decrease in PaCO2 occurs when the effective alveolar ventilation increases to a supernormal level. Important clinical examples of this decrease are seen in patients with tissue hypoxia (caused by heart failure, anemia, or residence at high altitudes), certain lung diseases, and enhanced central stimulation of respiration (resulting from such conditions as anxiety, pain, sepsis, pregnancy, neurologic disorders, hepatic failure, or salicylate intoxication). A decrease in PaCO2 (hypocapnia) elicits two responses with opposing effects on blood pH. In the short term, a decrease in PaCO2 alkalinizes extracellular fluid to the extent calculable by the Henderson–Hasselbalch equation (pH = 6.1 + log [bicarbonate/PaCO2 · 0.03]). Over a longer period (6 to 72 hours), however, renal acid excretion is inhibited (i.e., the body retains acid), resulting in a reduction in plasma bicarbonate that acidifies extracellular fluid and thereby corrects blood pH toward normal.
To diagnose respiratory alkalosis, reference data are needed that define the quantitative changes in the plasma bicarbonate and blood hydrogen concentrations or in pH in response to a given decrease in PaCO2 in normal subjects studied under conditions of strict metabolic control. By comparing a patient's blood gas values with such reference data, the clinician could identify chronic respiratory alkalosis that was present either as a single entity or along with coexisting acid–base disorders (mixed acid–base disturbances). In addition, accurate diagnostic criteria for chronic respiratory alkalosis would also enable the physician to gain important information about the time course of the underlying disease processes, such as fever, sepsis, or central nervous system or hepatic dysfunction and therefore to distinguish acute from chronic respiratory alkalosis. In contrast to the situation with other respiratory acid–base disturbances (acute and chronic respiratory acidosis and acute respiratory alkalosis1 2 3), reference data for the diagnosis of chronic respiratory alkalosis do not exist for humans because of the methodologic difficulty of studying normal human subjects in a hypoxic environment. Consequently, clinicians have been dependent on data obtained in studies of dogs.4
Stimulation of the medullary respiratory center in metabolic acidosis induces secondary hyperventilation, resulting in a decrease in PaCO2 (secondary hypocapnia). The effects of secondary hypocapnia on acid–base equilibrium in humans are currently unknown. It is assumed that the decrease in PaCO2 induced by secondary hyperventilation increases the blood pH toward normal, as predicted by the Henderson–Hasselbalch equation. On the basis of this premise, it has been recommended that critically ill patients with metabolic acidosis and ventilatory insufficiency be treated with mechanical hyperventilation.5 In dogs with metabolic acidosis (plasma bicarbonate concentration <18 mmol per liter), however, the reduction in renal acid excretion induced by the decrease in PaCO2 was so great that the plasma bicarbonate concentration decreased to a level that further lowered blood pH and thus paradoxically worsened the acidosis. This renal response to secondary hyperventilation has been called maladaptive, because the reduction in renal acid excretion (and thus in the plasma bicarbonate concentration) overrode the direct alkalinizing effect of the decrease in PaCO2.6 , 7
Whether and to what extent such a maladaptive renal response to hypocapnia occurs in humans is currently an unresolved clinical question of major importance. If the response occurs in humans, this maladaptive behavior would necessitate a reappraisal of the usefulness and effects of spontaneous and therapeutic hyperventilation in patients with clinical disorders of acid–base equilibrium. We undertook this study both to provide an experimental basis for the clinical diagnosis of chronic respiratory alkalosis in humans and to investigate whether a maladaptive renal response to a decrease in PaCO2 occurs in human metabolic acidosis. To this end, we determined the acid–base response to chronic hypocapnia (a long-term decrease in PaCO2) induced by hypobaric hypoxia in normal subjects with induced acidosis and similar subjects without acidosis under conditions of metabolic balance.
Methods
The protocol was designed to measure the renal and systemic acid–base response to chronic hypocapnia in normal subjects and acid-fed normal subjects. The acid–base and electrolyte composition of plasma and urine was determined during metabolicbalance studies in nine normal men (mean age [±SD], 28.1±3.6 years; weight, 75.4±9.3 kg). None were smokers or were taking any drugs before or during the study. They ate a constant metabolic diet for at least six days before the study (prestudy phase) and during all three study periods (control, hypocapnia, and recovery). The diet provided the following for every kilogram of body weight per day: 1.75 mmol of sodium, 1.53 mmol of potassium, 0.395 mmol (15.8 mg) of calcium, 0.51 mmol (15.8 rag) of phosphorus, 15.7 mmol (0.22 g) of nitrogen, 35 kcal, and 40.7 ml of water. Five of the nine subjects were fed 4.2 mmol of ammonium chloride per kilogram per day in gelatin capsules (divided into four doses of 1.05 mmol per kilogram each) throughout the study period. The initial dose, given six days before the study began, was 1 mmol per kilogram per day, and it was gradually increased to the full dose during the prestudy phase. The ammonium chloride was always taken after meals. Preliminary studies had established that ammonium chloride could be tolerated at this high dose without any side effects under these circumstances. Fasting arterialized venous blood samples8 were obtained in a heparin-coated syringe from a heated hand or forearm vein at 7 a.m., lO 1/2 hours after the last dose of ammonium chloride in the acid-fed group. Blood samples were accepted only if the partial pressure of oxygen was greater than 70 mm Hg (9.3 kPa). Weight and oral temperature were determined at the time of blood sampling; 24-hour urine samples were collected in plastic bottles containing mineral oil and thymol—chloroform. Exercise was limited to minimal ambulation throughout the study. During each study period, a subject was considered to be in a steady state when plasma values obtained on three consecutive days varied by no more than 1.5 mmol per liter for bicarbonate and bv no more than 3 mm Hg (0.4 kPa) for PaCO2.
All the subjects were medical students who volunteered for the study, were paid for their participation, and gave informed consent. The study protocol was approved by the ethics committee of the University of Berne School of Medicine.
Experimental Design
In order to establish a normal base line, the subjects were initially studied for four days (control period) in the hospital in Berne, at an altitude of 500 m above sea level (barometric pressure, 717 mm Hg; and partial pressure of oxygen, 150 mm Hg [20 kPa]).9
Altitude-induced hypobaric hypoxia was used as a stimulus for chronic hypocapnia in the normal subjects with acidosis induced by ammonium chloride and in those without acidosis. The subjects were taken by train to the High Alpine Research Station on the Jungfraujoch in the Swiss Alps, a well-equipped, heated research station at an elevation of 3450 m (barometric pressure, 497 mm Hg; and partial pressure of oxygen, 104 mm Hg [13.9 kPa]) that was transformed into a metabolic ward unit for this study. Transportation between the hospital and the Jungfraujoch took less than 3 1/2 hours. Observations were carried out for six days (hypocapnia period) at 3450 m, during which period a new steady state of acid–base equilibrium was established in all the subjects. All the subjects were then studied further during a recovery period of six days in Berne, at an elevation of 500 m.
Analytic Procedures
All measurements were performed in duplicate. The pH of blood and urine was determined anaerobically at 37°C with use of a Corning model 278 blood gas analyzer and a Corning model 250 pH meter (Ciba—Corning, Dietlikon, Switzerland). The total carbon dioxide content of plasma and urine was determined with a Corning model 965 carbon dioxide analyzer. Arterial blood carbon dioxide tension and plasma bicarbonate levels were calculated according to the Henderson–Hasselbalch equation with a pK′ of 6.10 and solubility coefficients of 0.0301 for plasma and 0.0309 for urine. The blood pH, pK′, and solubility of carbon dioxide were corrected for temperature,10 , 11 and the pK′ for urine was corrected for ionic strength.12 Sodium and potassium levels in plasma and urine were determined by flame photometry with lithium as an internal standard, chloride levels by the method of Cotlove and Nishi,13 and creatinine by a modification of the method of Jaffe.14 Plasma albumin levels were determined by the bromocresol-green method.15 Urinary ammonia was measured by the indophenol method.16 The concentration of titratable acid was determined in acidified urine by sodium hydroxide titration from urinary pH to blood pH. Net acid excretion was calculated as the sum of urinary ammonium and titratable acid minus urinary bicarbonate. Urinary cortisol and urinary aldosterone-18-glucuronide were measured with specific radioimmunoassays.17 , 18 Statistical significance was determined by linear regression analysis, covariance analysis, and Student's t-test for paired or unpaired data, as appropriate. All probability values presented for t-tests reflect two-tailed testing.
Results
Both sets of subjects tolerated the protocol well. The subjects with induced acidosis adapted to the high altitude without any symptoms, whereas the other subjects had moderate-to-severe headache and insomnia during the first night at high altitude. This observation, also made by Douglas et al.,19 may be of value in determining the mechanisms by which carbonic anhydrase inhibition prevents mountain sickness.20 , 21 All the subjects with acidosis lost weight during the prestudy phase (while receiving increasing doses of ammonium chloride), but they maintained a constant body weight during the study itself (Table 1Table 1
Effect of Chronic Hypocapnia on Steady-State Plasma Acid–Base and Electrolyte Composition in Normal Subjects and Normal Subjects with Induced Acidosis.*).The mean steady-state plasma acid–base and electrolyte composition, albumin concentration, creatinine clearance rate, and body weight for the two groups are shown in Table 1. The changes in steady-state plasma acid–base composition are shown in Figure 1Figure 1
Steady-State Plasma Acid–Base Indexes during the Control (C), Hypocapnia (H), and Recovery (R) Periods in Normal Subjects with and Those without Induced Acidosis.. The steady-state values for urinary acid–base and electrolyte excretion are given in Table 2Table 2
Effect of Chronic Hypocapnia on Urinary Acid–Base and Electrolyte Excretion in Normal Subjects and Normal Subjects with Induced Acidosis.*. The administration of ammonium chloride induced a moderately severe, chronic hyperchloremic metabolic acidosis in all five subjects who received it (Table 1, Fig. 1). The mean (±SE) steady-state plasma bicarbonate level was 13.2±0.5 mmol per liter lower in the acid-fed group than in the non-acid-fed group during the control period (P<0.001).Exposure to high altitude (a hypobaric, hypoxic environment) induced sustained hypocapnia in the normal subjects without induced acidosis and significantly exacerbated preexisting hypocapnia in the subjects with acidosis (Table 1, Fig. 1). In the subjects without acidosis, the mean PaCO2 decreased from 39.1 ±0.6 mm Hg (5.2±0.08 kPa) to 30.7±0.4 mm Hg (4.1 ±0.05 kPa; P<0.001); in the subjects with acidosis, the mean PaCO2 decreased from 24.7±0.9 mm Hg (3.3±12 kPa) to 20.2±0.7 mm Hg (2.7±0.9 kPa; P<0.005).
Effects of Chronic Hypocapnia on Plasma Acid–Base Composition and Urinary Acid–Base Excretion
The induction of chronic hypocapnia resulted in a sustained and significant reduction in the plasma bicarbonate concentration in both normal and acid-fed subjects. In both groups, the percent decrease in plasma bicarbonate was significantly less (ratio less than 1) than the corresponding decrease in PaCO2 (Fig. 2Figure 2
Decreases in Steady-State Plasma Bicarbonate Concentration and PaCO2 in Normal Subjects without Induced Acidosis (Open Circles) and Those with Acidosis (Solid Circles).). This resulted in a sustained alkalemic response (evidenced by a decrease in blood hydrogen ion concentrations or an increase in blood pH) (Fig. 1).Figure 3Figure 3
Relations of Blood Hydrogen Ion Concentration and Plasma Bicarbonate Concentration to PaCO2 in Normal Subjects with and Those without Induced Acidosis. shows that in the subjects both with and without induced acidosis, significant linear relations were obtained for the regression of the steady-state mean plasma bicarbonate level and the blood hydrogen ion concentration on PaCO2 for the range of PaCO2 values observed. The slopes of the regression lines for the hydrogen ion concentration on PaCO2 and the bicarbonate concentration on PaCO2 were nearly identical in the subjects with acidosis and those without. (Figure 3 also shows the 95 percent confidence limits for normal subjects with uncomplicated chronic respiratory alkalosis.)In the normal subjects, net acid excretion decreased sharply after the induction of hypocapnia (Table 2) and returned to control values by the end of the period of hypocapnia. The hypocapnia-induced decrease in net acid excretion was largely the result of bicarbonaturia and a decrease in the excretion of titratable acid. The cumulative decrement during the hypocapnic period (–88 mmol, P<0.001) was nearly reversed ( + 69 mmol, P<0.001) during the recovery period. In the acid-fed subjects, we detected no significant differences in daily net excretion of acid in the urine or in the cumulative change in net acid excretion. On the basis of a space of distribution (volume in which bicarbonate apparently distributes) of 50 percent of body weight for bicarbonate, the small hypocapniainduced decrease in the plasma bicarbonate level in the acid-fed subjects would be expected to result from a cumulative decrement in net acid excretion of only 30 to 40 mmol for the entire period of hypocapnia. It is probably impossible to detect such small changes because of the relatively large mean daily net acid excretion in subjects eating large acid loads.6 , 7 More important, the nearly identical rates of net acid excretion in the subjects with acidosis during the three steady-state periods (332, 336, and 333 mmol per 24 hours) exclude the possibility of a substantial hypocapnia-induced extrarenal release of acid or base into the extracellular fluid (for example, increased release of bicarbonate from bone buffers or gastrointestinal mal-absorption of acid during hypocapnia). Thus, in the steady state of hypocapnia in both groups (days 4 through 6 of the hypocapnia period), chronic hypocapnia reduced the set-point for the regulation of the plasma bicarbonate concentration by the kidney.
Effects of Hypocapnia on Plasma Electrolyte Composition and Urinary Electrolyte Excretion
The plasma potassium concentrations decreased promptly in response to hypocapnia in both the subjects with induced acidosis and those without acidosis. The decrease was sustained in both groups and reversed (although only partially in the acid-fed subjects) during recovery. Urinary potassium excretion decreased significantly during the first three days of hypocapnia (cumulative change in excretion: –49 mmol in the subjects without acidosis, P<0.001; and –92 mmol in the subjects with induced acidosis, P<0.001) and returned to values that did not differ significantly from the control values at the end of the period of hypocapnia (Table 2). Urinary aldosterone excretion averaged 32±2 nmol (12±1 μg) per 24 hours in the normal subjects during the control period and had not changed significantly by days 5 through 6 of the hypocapnia period (28±4 nmol [10±1 μg] per 24 hours). In the subjects with acidosis, aldosterone excretion during the control period averaged 165±43 nmol (60±16 μg) per 24 hours and was not significantly different by days 4 through 6 of hypocapnia (96±18 nmol [35±7 μg] per 24 hours). The aldosterone-excretion values were significantly higher in the acid-fed subjects during both the control period (P<0.0025) and the hypocapnia period (P<0.001). Evidence for augmented aldosterone secretion in response to long-term ammonium chloride loading has been reported by Schambelan et al.22 Urinary cortisol excretion averaged 75±13 nmol (28±5 μg) per 24 hours during the control period for the normal subjects and did not change significantly during hypocapnia (112±27 nmol [41±10 μg] per 24 hours). In the acid-fed subjects, the cortisol excretion during the control period averaged 133±35 nmol (49± 13 μg) per 24 hours and was not significantly different during the period of hypocapnia (130± 17 nmol [48±6 g] per 24 hours). No significant differences were found for cortisol excretion between normal and acid-fed subjects during either period.
Hypocapnia did not result in significant changes in the plasma lactate concentration (measured on the last two days of each study period). Hematocrit values were also normal and did not change significantly during the study.
Discussion
This study addressed two important clinical questions. First, from a diagnostic standpoint, we defined steady-state plasma acid–base composition in humans with chronic respiratory alkalosis, thus providing an experimental basis for the diagnosis of the disorder as a single acid–base disturbance and for its identification in mixed disorders. Second, from a therapeutic perspective, we demonstrated that in chronic metabolic acidosis in humans, the hypocapnia-induced decrease in renal acid excretion and the resulting decrease in the plasma bicarbonate concentration are too small to override the direct alkalemic effect of the decrease in PaCO2.
A decrease in PaCO2 (hypocapnia) has two opposite effects on acid–base equilibrium. In the short term, the decrease in PaCO2 increases blood pH as predicted by the Henderson–Hasselbalch equation; in the longer term (after 6 to 72 hours), renal acid excretion is inhibited (i.e., the body retains acid), with a resulting decrease in the plasma bicarbonate concentration and blood pH.23 , 24 The relative importance of each of these two effects therefore determines the quantitative changes in the blood hydrogen ion concentration (or pH) and the plasma bicarbonate concentration in response to a given long-term decrease in PaCO2. If these interrelations are determined experimentally, diagnostic criteria for chronic respiratory alkalosis can be established. In contrast to the primary hypocapnia that characterizes uncomplicated chronic respiratory alkalosis, secondary hypocapnia is induced by hyperventilation in response to metabolic acidosis. In this disorder, acid excretion is stimulated by the low blood pH, the low plasma bicarbonate concentration, or both, but it is unknown whether and to what extent the secondary decrease in PaCO2 affects renal acid excretion and thus plasma bicarbonate concentration and blood pH in humans. A persistent decrease in the plasma bicarbonate concentration, caused by the kidney's response to secondary hypocapnia, would, of course, be counterproductive to the correction of acidemia in metabolic acidosis.
In studies in dogs, hypocapnia caused the kidney to decrease the plasma bicarbonate concentration (renal acid retention) even in the context of metabolic acidosis — that is, when the plasma bicarbonate concentration was already low.6 , 7 Hypocapnia in dogs induced a decrease in plasma bicarbonate that was strictly proportional to the decrease in PaCO2 over a wide range of values for PaCO2 and plasma bicarbonate concentration (ratio of the change in the bicarbonate concentration to the change in PaCO2, 0.54 mmol per liter per 1 mm Hg [0.13 kPa]).4 , 6 , 7 Because of this proportionality, the percent change induced in the plasma bicarbonate concentration by hypocapnia became greater than the percent change in PaCO2 at the lower plasma bicarbonate concentrations of metabolic acidosis (bicarbonate concentration, <18 mmol per liter). This relation compelled a decrease in blood pH (an increase in the hydrogen ion concentration) because of the mathematical relation among the bicarbonate concentration, PaCO2, and the hydrogen ion concentration, which is defined by the Henderson–Hasselbalch equation (the hydrogen ion concentration is proportional to the ratio of PaCO2 to the bicarbonate concentration). These studies in dogs raised two sets of intriguing clinical questions. First, might spontaneous or therapeutic mechanical ventilation worsen acidemia in patients with metabolic acidosis? Second, might a patient with metabolic acidosis and worsening acidemia have not only worsening metabolic acidosis or superimposed respiratory acidosis but, alternatively, coexisting chronic respiratory alkalosis?
Our findings demonstrate that in normal subjects, chronic hypocapnia decreases the plasma bicarbonate concentration when the initial plasma bicarbonate concentration is normal and also when the initial concentration is low (in cases of metabolic acidosis). However, the percent decrease in plasma bicarbonate was significantly smaller than the percent decrease in PaCO2 (Fig. 2). Therefore, in sharp contrast to the observations made in dogs, blood pH increased, and this alkalemic response was of similar magnitude in both the subjects with acidosis and those without (Fig. 3). The finding that the relation of steady-state plasma bicarbonate concentration to PaCO2 was very similar in the subjects with uncomplicated chronic respiratory alkalosis (0.41) and the subjects with chronic respiratory alkalosis superimposed on metabolic acidosis (0.42) indicates that the kidney reacts to changes in PaCO2 regardless of blood pH or plasma bicarbonate concentration. The decrease in PaCO2 thus inhibited renal acid excretion even in our subjects with moderately severe acidosis. However, in contrast to the effect in dogs, the inhibitory effect of decreased PaCO2 on renal acid excretion in subjects with acidosis is too small to abolish the direct alkalemic effect of the decrease in PaCO2. For reasons of safety, we did not attempt to induce more severe acidosis in our subjects (initial steady-state plasma bicarbonate concentration below 10 mmol per liter). From the current data, therefore, it is impossible to predict the slope of line for the relation of the bicarbonate concentration on PaCO2 at very low initial plasma bicarbonate concentrations.
These results thus indicate that hypocapnia has a net alkalemic, homeostatic effect in humans with metabolic acidosis. The clinical consequence of this finding is that worsening acidemia in patients with metabolic acidosis cannot be ascribed to developing or coexisting respiratory alkalosis. Rather, hyperventilation can be expected to enhance the return of blood pH toward normal values in metabolic acidosis.
Mineral acid feeding (administration of ammonium chloride) was chosen as the means to produce stable metabolic acidosis in this study. Since the addition of 1 mmol of acid consumes 1 mmol of base, the addition of acid (ammonium chloride or overproduced lactic acid or keto acids) results in the same effects on acid–base equilibrium as the removal of base from the body (for instance, by diarrhea or pancreatic drainage). Therefore, although the ingestion of mineral acid is an unusual cause of clinical metabolic acidosis, our use of this model of the disorder resulted in a pattern of plasma acid–base composition that was indistinguishable from that found in more common but less well controlled models.
These studies also provide diagnostic criteria for identifying chronic respiratory alkalosis in humans. These criteria are defined by confidence limits indicating the expected changes in the plasma bicarbonate concentration and blood hydrogen ion concentration (pH) for a given decrease in PaCO2 (Fig. 3). Previously, the diagnosis of this disorder was based largely on the confidence limits established in dogs.4 Several studies of humans with chronic hypocapnia have been performed, from which an exceedingly wide range of values for the ratio of the change in the hydrogen concentration to the change in PaCO2, from 0.25 to 0.49 nmol per liter per 1 mm Hg (0.13 kPa) can be calculated.21 , 25 26 27 28 29 Unfortunately, each of these studies has been difficult to accept as representative of uncomplicated chronic respiratory alkalosis, either because the study conditions did not permit the exclusion of metabolic acid–base disorders (i.e., those caused by associated disease states, exercise, lack of a fixed diet, or immobility) or because the investigators did not document steady-state conditions or obtained their results by retrospective analysis.
The present results in normal subjects both with and without induced acidosis demonstrate that chronic hypocapnia causes hypokalemia despite a transient early decrease in renal potassium excretion. The acute hypokalemic response to hypocapnia (reviewed by Adrogué and Madias30) is associated with both transient kaliuresis (lasting for hours)31 32 33 and a shift of potassium from the extracellular to the intracellular space.34 , 35 The decrease in urinary potassium excretion during the first three days of hypocapnia (see the Results section) is consistent with the known modulation of daily potassium excretion by the plasma potassium concentration.36 The return of potassium excretion to control levels by day 6 of hypocapnia, despite sustained hypokalemia and in the absence of evidence of increased aldosterone or cortisol secretion, suggests, however, that chronic hypocapnia decreased the set-point for the renal regulation of the plasma potassium concentration — that is, it induced renal potassium wasting (Tables 1 and 2). The mechanism of this apparent hypocapnia-induced renal potassium wasting is presently unknown.
Despite the limitations inherent in our small sample and the homogeneity of study population, this study provides strong evidence that chronic hypocapnia induces a significant and sustained increase in blood pH (alkalemic response) in subjects with chronic extrarenal metabolic acidosis. This alkalemic response is similar in magnitude to that which occurs in normal subjects. In both normal subjects with acidosis and those without, the hypocapnia-induced suppression of renal acid excretion was too small to override the direct alkalemic effect of hypocapnia. Thus, clinically, worsening acidemia in a patient with metabolic acidosis indicates worsening of the metabolic disturbance or additional respiratory acidosis and not coexisting respiratory alkalosis. Chronic hypocapnia, either spontaneous or therapeutically instituted in cases of insufficient or failing ventilatory response, enhances the homeostatic return of blood pH toward normal values in metabolic acidosis. Finally, the reference data we have derived for the diagnosis of chronic respiratory alkalosis will permit more accurate and rapid identification of this common clinical disorder.
Supported by a grant from the Bernische Hochschulstiftung and by the Department of Medicine, Insel University Hospital.
We are indebted to Ciba—Corning (Dietlikon, Switzerland) for the equipment for blood gas analysis, to Viollier Laboratories (Basel and Berne, Switzerland) for performing the aldosterone and cortisol radioimmunoassays, and to Hausmann Laboratories (St. Gallen, Switzerland) for supplying the pH titrator and chemicals.
Source Information
From the Department of Medicine, Insel University Hospital, Berne, Switzerland (R.K., I.B., D.H.), and the Division of Nephrology, Department of Medicine, San Francisco General Hospital, University of California, San Francisco (H.N.H). Address reprint requests to Dr. Krapf at the Department of Medicine, Insel University Hospital, CH–3010 Berne, Switzerland.
- References
References
1
Brackett
Full Text | Web of Science | Medline2
Brackett
Full Text | Web of Science | Medline3
Arbus
Full Text | Web of Science | Medline4
Gennari
CrossRef | Web of Science | Medline5
Fishman AP. Pulmonary diseases and disorders. 2nd ed. Vol. 3. New York: McGraw-Hill, 1988
6
Cohen
CrossRef | Web of Science | Medline7
Madias
CrossRef | Web of Science | Medline8
Forster
Web of Science | Medline9
Altman PL, Dittmer DS. Respiration and circulation. Bethcsda. Md.: Federation of American Societies for Experimental Biology, 1971:12
10
Severinghaus
Web of Science | Medline11
Rosenthal
Web of Science | Medline12
Sendroy
Web of Science13
Cotlove
Web of Science | Medline14
Helger
Medline15
Duggan
Web of Science | Medline16
Logsdon
CrossRef | Web of Science | Medline17
Foster
Web of Science | Medline18
Ogihara
CrossRef | Web of Science | Medline19
Douglas
Medline20
Dickinson
CrossRef | Web of Science | Medline21
Forwand
Full Text | Web of Science | Medline22
Schambelan
Web of Science | Medline23
Brazeau
Web of Science | Medline24
Relman
CrossRef | Web of Science | Medline25
Severinghaus
Web of Science | Medline26
Forster
Web of Science | Medline27
Dempsey
CrossRef | Medline28
Dempsey
CrossRef | Web of Science | Medline29
Grimbert
Medline30
Adrogué
CrossRef | Web of Science | Medline31
Gledhill
CrossRef | Web of Science | Medline32
Mostellar
CrossRef | Web of Science | Medline33
Barker
CrossRef | Web of Science | Medline34
Giebisch
CrossRef | Web of Science | Medline35
Burnell
CrossRef | Web of Science | Medline36
Young
Web of Science | Medline
- Citing Articles (17)
Citing Articles
1
Horacio J Adrogué, F John Gennari, John H Galla, Nicolaos E Madias. (2009) Assessing acid–base disorders. Kidney International 76:12, 1239-1247
CrossRef2
(2009) Blood Oxygen on Mount Everest. New England Journal of Medicine 360:18, 1908-1910
Full Text3
Pablo Alvarez Maldonado, Ulises Cerón Díaz, Alfredo Sierra Unzueta. (2009) SmartCare™ closed-loop system and the altitude problem. Intensive Care Medicine 35:4, 755-755
CrossRef4
Sumit Mohan, Manasvi Jaitly, Constance M. Park, Jen-Tse Cheng, Velvie A. Pogue. (2008) In Reply. American Journal of Kidney Diseases 52:1, 196-197
CrossRef5
C. G. Morris, J. Low. (2008) Metabolic acidosis in the critically ill: Part 1. Classification and pathophysiology. Anaesthesia 63:3, 294-301
CrossRef6
2008. Gas Exchange and Acid-Base Physiology. , 179-200.
CrossRef7
Shubhada N. Ahya, Maria José Soler, Josh Levitsky, Daniel Batlle. (2006) Acid-Base and Potassium Disorders in Liver Disease. Seminars in Nephrology 26:6, 466-470
CrossRef8
Henry N. Hulter, Reto Krapf. (2006) Interrelationships Among Hypoxia-Inducible Factor Biology and Acid-Base Equilibrium. Seminars in Nephrology 26:6, 454-465
CrossRef9
O. Witzke, U. Heemann. (2006) Diagnose und Therapie von Störungen des Säure-Basen-Haushaltes. Der Nephrologe 1:2, 113-126
CrossRef10
Ri-Li Ge, Tony G. Babb, Mark Sivieri, Geir K. Resaland, Trine Karlsen, Jim Stray-Gundersen, Benjamin D. Levine. (2006) Urine Acid–Base Compensation at Simulated Moderate Altitude. High Altitude Medicine & Biology 7:1, 64-71
CrossRef11
(2002) Sightings. High Altitude Medicine & Biology 3:2, 149-155
CrossRef12
Katia Mahlbacher, Anita Sicuro, Hans Gerber, Henry N. Hulter, Reto Krapf. (1999) Growth hormone corrects acidosis-induced renal nitrogen wasting and renal phosphate depletion and attenuates renal magnesium wasting in humans. Metabolism 48:6, 763-770
CrossRef13
Niels Vidiendal Olsen, Henrik Christensen, Tom Klausen, Niels Fogh-Andersen, Inger Plum, Inge-Lis Kanstrup, Jesper Melchior Hansen. (1998) Effects of Hyperventilation and Hypocapnic/Normocapnic Hypoxemia on Renal Function and Lithium Clearance in Humans. Anesthesiology 89:6, 1389-1400
CrossRef14
Adrogué, Horacio J., Madias, Nicolaos E., . (1998) Management of Life-Threatening Acid–Base Disorders. New England Journal of Medicine 338:2, 107-111
Full Text15
Marco Brüngger, Henry N Hulter, Reto Krapf. (1997) Effect of chronic metabolic acidosis on the growth hormone/IGF-1 endocrine axis: New cause of growth hormone insensitivity in humans. Kidney International 51:1, 216-221
CrossRef16
Reto Krapf, Pia Caduff, Philippe Wagdi, Max Stäubli, Henry N Hulter. (1995) Plasma potassium response to acute respiratory alkalosis. Kidney International 47:1, 217-224
CrossRef17
Reto Krapf, Philippe Jaeger, Henry N Hulter, C Fehlman, R Takkinen. (1992) Chronic respiratory alkalosis induces renal PTH-resistance, hyperphosphatemia and hypocalcemia in humans. Kidney International 42:3, 727-734
CrossRef
