Hypoxemia during hemodialysis: a critical review of the facts

Hidden risks associated with conventional short intermittent hemodialysis: A call for action to mitigate cardiovascular risk and morbidity

The development of maintenance hemodialysis (HD) for end stage kidney disease patients is a success story that continues to save many lives. Nevertheless, intermittent renal replacement therapy is also a source of recurrent stress for patients. Conventional thrice weekly short HD is an imperfect treatment that only partially corrects uremic abnormalities, increases cardiovascular risk, and exacerbates disease burden. Altering cycles of fluid loading associated with cardiac stretching (interdialytic phase) and then fluid unloading (intradialytic phase) likely contribute to cardiac and vascular damage. This unphysiologic treatment profile combined with cyclic disturbances including osmotic and electrolytic shifts may contribute to morbidity in dialysis patients and augment the health burden of treatment. As such, HD patients are exposed to multiple stressors including cardiocirculatory, inflammatory, biologic, hypoxemic, and nutritional. This cascade of events can be termed the dialysis stress storm and sickness syndrome. Mitigating cardiovascular risk and morbidity associated with conventional intermittent HD appears to be a priority for improving patient experience and reducing disease burden. In this in-depth review, we summarize the hidden effects of intermittent HD therapy, and call for action to improve delivered HD and develop treatment schedules that are better tolerated and associated with fewer adverse effects.

Dialysis-Induced Cardiovascular and Multiorgan Morbidity

Hemodialysis has saved many lives, albeit with significant residual mortality. Although poor outcomes may reflect advanced age and comorbid conditions, hemodialysis per se may harm patients, contributing to morbidity and perhaps mortality. Systemic circulatory "stress" resulting from hemodialysis treatment schedule may act as a disease modifier, resulting in a multiorgan injury superimposed on preexistent comorbidities. New functional intradialytic imaging (i.e., echocardiography, cardiac magnetic resonance imaging [MRI]) and kinetic of specific cardiac biomarkers (i.e., Troponin I) have clearly documented this additional source of end-organ damage. In this context, several factors resulting from patient-hemodialysis interaction and/or patient management have been identified. Intradialytic hypovolemia, hypotensive episodes, hypoxemia, solutes, and electrolyte fluxes as well as cardiac arrhythmias are among the contributing factors to systemic circulatory stress that are induced by hemodialysis. Additionally, these factors contribute to patients' symptom burden, impair cognitive function, and finally have a negative impact on patients' perception and quality of life. In this review, we summarize the adverse systemic effects of current intermittent hemodialysis therapy, their pathophysiologic consequences, review the evidence for interventions that are cardioprotective, and explore new approaches that may further reduce the systemic burden of hemodialysis. These include improved biocompatible materials, smart dialysis machines that automatically may control the fluxes of solutes and electrolytes, volume and hemodynamic control, health trackers, and potentially disruptive technologies facilitating a more personalized medicine approach.

Influence of Acetate and Bicarbonate Hemodialysis on the Plasma Erythropoietin Concentration in Patients with Chronic Renal Failure

The present study aimed to assess the influence of acetate (AC) and bicarbonate (BI) hemodialysis with a cuprophane membrane on plasma erythropoietin (EPO) levels in 30 patients with chronic renal failure (CRF). Fifteen patients were severely anemic with an Hct below 30%, while 13 were moderately anemic with an Hct equal to or more than 30%. None had received rHuEPO or ACE inhibitors before the study. Blood samples for EPO and pO2 estimation were withdrawn before and after 1, 2 and 5 h of AC dialysis. In 11 of the patients with an Hct < 30% the same protocol was repeated during BI dialysis. In spite of a significant decrease in pO2 during the first hour of AC dialysis, the plasma EPO concentration decreased significnatly in both severely and moderately anemic CRF patients from 35.4 ± 2.6 to 23.6 ± 3.5 mU/ml after 5 h of hemodialysis (p = 0.0001 by paired t-test) and from 39.5 ± 8.6 to 27.5 ± 8.9 mil/ml (p < 0.01) respectively. In contrast to AC dialysis, Bi dialysis failed to change the plasma EPO concentration (from 32.3 ± 4.1 to 30.3 ± 4.7mil/ml after 5 h of hemodialysis). Conclusions AC but not BI hemodialysis shows a significant suppressive effect on plasma EPO levels in anemic CRF patients in spite of a significant decline in pO2. Factors other than pO2 seem to be involved in the regulation of EPO secretion in CRF patients during AC dialysis.

Acid-base alterations in ESRD and effects of hemodialysis

Acid-base alterations in patients with kidney failure and on hemodialysis (HD) treatment contribute to (1) intradialytic hypercapnia and hypoxia, (2) hemodynamic instability and cardiac arrhythmia, (3) systemic inflammation, and (4) a number of associated electrolyte alterations including potentiating effects of hypokalemia, hypocalcemia and, chronically, soft-tissue and vascular calcification, imparting poor prognosis and mortality. This paper discusses acid-base regulation and pathogenesis of dysregulation in patients with kidney failure. Major organ and systemic effects of acid-base perturbations with a specific focus on kidney failure patients on HD are emphasized, and potential mitigating strategies proposed. The high rate of HD-related complications, specifically those that can be accounted for by rapid and steep acid-base perturbations imposed by HD treatment, attests to the pressing need for investigations to establish a better dialysis regimen.

Intradialytic Hypoxemia and Clinical Outcomes in Patients on Hemodialysis

BACKGROUND AND OBJECTIVES

Intradialytic hypoxemia has been recognized for decades, but its associations with outcomes have not yet been assessed in a large patient cohort.

DESIGN, SETTING, PARTICIPANTS, & MEASUREMENTS

Our retrospective cohort study was conducted between January of 2012 and January of 2015. We recorded blood oxygen saturation every minute during hemodialysis in patients with arteriovenous access. A 6-month baseline period with at least 10 treatments with oxygen saturation measurements preceded a 12-month follow-up. Patients were stratified by the presence or absence of prolonged intradialytic hypoxemia defined as oxygen saturation <90% for at least one third of the treatment time. Demographic, laboratory, and treatment data and hospitalization and mortality rates were compared between the groups. Multivariate Cox regression analysis was used to assess baseline predictors of all-cause mortality during follow-up.

RESULTS

In total, 100 (10%) of 983 patients had prolonged intradialytic hypoxemia. These patients were older (+3.6 years; 95% confidence interval, 0.8 to 6.3), had longer dialysis vintage (+1.2 years; 95% confidence interval, 0.3 to 2.1), and had higher prevalence of congestive heart failure (+10.8%; 95% confidence interval, 1.6 to 20.7) and chronic obstructive pulmonary disease (+13%; 95% confidence interval, 5 to 21.2). They also resembled an inflammatory phenotype, with lower serum albumin levels (-0.1 g/dl; 95% confidence interval, -0.2 to 0) and higher neutrophil-to-lymphocyte ratios (+1; 95% confidence interval, 0.5 to 1.6). They had lower hemoglobin levels (-0.2 g/dl; 95% confidence interval, -0.4 to 0) and required more erythropoietin (+1374 U per hemodialysis treatment; 95% confidence interval, 343 to 2405). During follow-up, all-cause hospitalization (1113 hospitalizations; univariate hazard ratio, 1.46; 95% confidence interval, 1.22 to 1.73) and mortality (89 deaths; adjusted hazard ratio, 1.98; 95% confidence interval, 1.14 to 3.43) were higher in patients with prolonged intradialytic hypoxemia.

CONCLUSIONS

Prolonged intradialytic hypoxemia was associated with laboratory indicators of inflammation, higher erythropoietin requirements, and higher all-cause hospitalization and mortality.

Possible application of hyperbaric oxygen technology in the management of urogenital and renal diseases

The purpose of this report is to explore possible therapeutic use of hyperbaric oxygen (HBO(2)) technology on renal and urogenital diseases. HBO(2) reduces inflammation, immunity and inflammatory cytokines, stimulates wound repair and angioneogenesis, maintains tissue oxygenation, increases antioxidant enzymes and heals tissue hypoxia and radionecrosis. A literature review of peer-reviewed articles that address HBO(2), genitourological diseases, renal disease, and dialysis was performed. The paper reviews complications of renal diseases, dialysis, clinical applications of HBO(2), and effect of HBO(2) on renal and urogenital diseases. HBO(2) was used successfully to treat calcific uraemic arteriolopathy, and in many cases of acute renal failure. This technique is particularly useful in the treatment of intractable haemorrhagic cystitis secondary to pelvic radiation therapy and Fournier's gangrene. Clearly HBO(2) might play a role in the management of urogenital diseases, urinary bladder dysfunction and diseases, testicular pathology, renal diseases, and post-traumatic ischaemic injury and/or impaired wound healing and infections. The possible role of HBO(2) for autoimmune diseases, uraemic osteodystrophy or neuropathy due to chronic renal diseases is discussed. The clinical application of this technology is expanding and the various biological influences of HBO(2) encourage testing its possible benefit in renal and urological diseases.

Shortness of breath during dialysis--a role of bicarbonate in dialysis fluid?

Occasionally dialysis patients show symptoms that indicate intolerance in the way dialysis is delivered. This paper describes two cases of transient shortness of breath during the initial treatments after starting online haemodiafiltration (HDF). Our actions to deal with these symptoms focused on reducing the bicarbonate gain during the first phase of the dialysis treatment. As the symptoms gradually disappeared we hypothesise that the bicarbonate concentration in the dialysis fluid may play an important role for the development of shortness of breath and hypoxemia during HDF treatments.

Evaluating the changes in alveolar permeability and lung ventilation in patients with chronic renal failure after haemodialysis using 99mTc-DTPA radioaerosol inhalation lung scan

Hypoxaemia occurring in patients with chronic renal failure (CRF) during haemodialysis (HD) has long been known. Several mechanisms of pathogenesis have been proposed. Before and after regular HD, lung ventilation (LV) and alveolar permeability (AP) were measured in 24 male patients with CRF (age, 61-75 years). LV and AP were determined by 99mTc-diethylenetriaminepentaacetic acid (99mTc-DTPA) radioaerosol inhalation lung scan (99mTc-DTPA lung scan). The LV images were visually interpreted according to established criteria, including the presence or absence of an inhomogeneous distribution, inverted base to apex gradient and segmental hypoventilation. The degree of AP in the total right lung was presented as the clearance rate (K; %.min-1) of the time-activity curve from dynamic total right lung images. Ten male normal controls (age, 62-76 years) were enrolled in the study for comparison. Ten of 24 (42%) cases showed an inhomogeneous distribution and eight of 24 (33%) cases showed hypoventilation on equilibrium LV images. After regular HD for 5 h, no significant changes in the LV images were found. Before HD, the CRF patient group (K=1.14+/-0.36%.min-1) had a significantly faster clearance rate than that of normal controls (0.75+/-0.14%.min-1, P<0.05). Moreover, the clearance rate after HD was significantly slower (0.87+/-0.15%.min-1, P<0.05) than that before HD. CRF can predispose patients to LV change and AP damage. After HD, the damage to AP is significantly improved. However, after HD, the change in LV is not significant.

Pulmonary infectious mortality among patients with end-stage renal disease

BACKGROUND

Infection is the second-leading cause of death among patients with end-stage renal disease (ESRD). This is due in part to advanced age, comorbid conditions, and immune dysfunction observed in uremic states. Although one may hypothesize that pulmonary infectious mortality is higher among patients with ESRD compared with the general population (GP), no such data are currently available.

METHODS

We compared annual pulmonary infectious mortality rates among patients with ESRD to those in the GP. The data were abstracted from the United States Renal Data System and the National Center for Health Statistics, respectively, and were stratified by age, gender, race, and presence or absence of diabetes mellitus (DM). In the GP, primary and multiple cause-of-death analyses were performed to account for potential limitations of the data sources.

RESULTS

Overall, pulmonary infectious mortality rate was 14-fold to 16-fold higher in dialysis patients and approximately twofold higher in renal transplant recipients compared with the GP. After stratification for age, differences between groups decreased but retained their magnitude.

CONCLUSION

Patients with ESRD treated with dialysis have higher pulmonary infectious mortality rates compared with the GP, even after stratification for age, race, and DM. Consequently, this patient population must be considered at high risk for the development of lethal pulmonary infections.

Cardiopulmonary events during hemodialysis: effects of dialysis membranes and dialysate buffers

Adverse cardiac and pulmonary events are frequently observed during hemodialysis and contribute to significant morbidity and mortality. The temporal relationship between these events during the intradialytic period has not been well defined. To examine the event rate and timing of silent ischemia, cardiac ectopy, and hypoxemia, we conducted a prospective, single-blind, randomized study of 10 subjects undergoing maintenance hemodialysis with four contiguous combinations of dialysis membranes (cuprammonium or polysulfone) and dialysates (acetate or bicarbonate). The frequency of oxygen desaturation events peaked during the first 2 hours, whereas silent myocardial ischemia and supraventricular ectopies occurred more often in the later hours. Ventricular ectopy occurred steadily throughout the intradialytic period. The combination of acetate dialysis and cuprammonium membrane is associated with the most frequent events. We conclude that cardiopulmonary events can occur frequently during hemodialysis, and the frequency is dependent on the type of dialysis membrane and dialysate buffer used.

Dissociation between complement activation, integrin expression and neutropenia during hemodialysis

Complement activation, neutrophil stimulation, increased cellular adhesiveness, transient leukocyte margination and pulmonary leukostasis take place during hemodialysis with cellulosic dialysis membranes. Several investigators have hypothesized that complement activation is primarily responsible for the acute neutropenia occurring during the early phase of bio-incompatible hemodialysis. We have investigated the relationship between complement activation, levels of expression of CD11b and CD61 integrins on neutrophils and platelets, neutrophil counts and blood gas measurements in patients dialyzed with three types of membranes, known to activate the complement system to a different extent. Polysulfone, cellulose acetate and cuprophane membranes were used subsequently in six patients in a prospective cross-over trial design to reduce inter-individual variability. Increased levels of CD61 and CD11b, as well as neutropenia, were detected regardless of the type of membrane used. We observed a high inter-individual variation with regard to complement activation suggesting varying susceptibility to dialysis membranes. We also report that the kinetics of anaphylatoxin generation were dissociated from those of the upregulation of adhesion molecules, early neutrophil margination and decrease in PaO2 during the first 30 min of hemodialysis. Similar results were obtained with all three types of dialysis membranes. The data strengthen the hypothesis that factors other than complement are involved in the induction of dialysis-related neutropenia and hypoxemia.

[Cardiac pathology of extracardiac origin (IX)> Cardiac pathology in the patient with chronic nephropathy]

Cardiac disease constitutes a common complication among patients with renal failure. This is partly due to the high incidence of shared risk factors, such as hypertension or diabetes mellitus, and some to specific factors inherent in renal disease. It implies a high incidence of cardiac failure and ischemic heart disease (frequently without significant coronary artery obstructions) with important associated morbidity and mortality. Pericardial disease, valvular involvement and arrhythmia are also common among these patients. The management of these complications in patients with endstage renal disease has some particularities, specially in the field of drug therapy.

Oxygen, arterial blood gases and ventilation are unchanged during dialysis in patients receiving pressure support ventilation

This study was undertaken to observe whether dialysis-induced alveolar hypoventilation and arterial hypoxaemia occur during bicarbonate haemodialysis in patients receiving partial mechanical support with pressure support ventilation. Nineteen patients admitted to the medical intensive care unit requiring mechanical ventilation and haemodialysis were enrolled. Arterial blood gas, white blood cell (WBC) count, minute ventilation, respiratory rate, breathing pattern and blood pressure were measured according to the following time schedule: pre-dialysis (time 0), and at 15, 30, 60, 120, 180, 240 min thereafter. Results showed that, with the use of cuprammonium dialyser, the WBC count dropped immediately and reached the nadir 15 min after haemodialysis. Thereafter, it recovered and overshot the pre-dialysis values until the end of dialysis. The bicarbonate dialysate indeed resulted in rapid and significant metabolic alkalosis. However, no decrease of PaO2 occurred throughout haemodialysis. The tidal volume, minute ventilation and breathing pattern remained stable during haemodialysis. We conclude that neither dialysis-induced alveolar hypoventilation nor arterial hypoxaemia developed during bicarbonate dialysis in patients mechanically ventilated with the pressure support ventilation.

Respiratory drive and pulmonary mechanics during haemodialysis with ultrafiltration in ventilated patients

The improvements of respiratory drive and pulmonary mechanics which follow haemodialysis with ultrafiltration in mechanically ventilated renal failure patients seem predictable but have not been studied before. In this study, 14 renal failure patients with stable haemodynamics mechanically ventilated with pressure support ventilation (PSV) were enrolled. Respiratory drive (represented as P0.1), pulmonary mechanics, breathing pattern, arterial blood gas and haemodynamics were measured according to the time schedule: pre-dialysis (Time 0), and at 60, 120, 180, 240 minutes thereafter. Following the removal of excess lung water during haemodialysis, auto-PEEP and patient's work of breathing (WOBp) decreased gradually. P0.1 lessened progressively along with the improvement in pulmonary mechanics. The changes in auto-PEEP and WOBp correlated closely to the pre- and post-dialysis decline of P0.1 (delta P0.1). There was a negative, moderately significant correlation between the amount of fluid ultrafiltrated during dialysis (delta UF) and the delta P0.1 (R = -0.54). The breathing pattern remained stable during dialysis. No hypoventilation or hypoxaemia occurred despite the development of metabolic alkalosis induced by bicarbonate dialysate. We have shown that respiratory drive decreases gradually during bicarbonate haemodialysis. The improvements of pulmonary mechanics, rather than the rapid alkalization of body fluids, responds to the decrease of P0.1 in renal failure patients ventilated with PSV.

Cardiac disease in chronic uremia: management

Heart disease is a common cause of morbidity in end-stage renal disease (ESRD) patients. The management of heart disease in these patients requires a multidimensional approach to the management of heart failure, coronary disease, and arrhythmias, and to risk factors such as hypertension, anemia, secondary hyperparathyroidism, and electrolyte/acid-base disturbances. Coronary artery disease management includes use of antianginal drugs and revascularization of coronary arteries with angioplasty +/- stent placement or coronary artery bypass grafting. The long-term outcomes of these procedures need to be assessed and improved. Hypertension occupies a major role in the pathogenesis of heart disease in ESRD, and early and adequate control of hypertension is likely to have a major impact on the progression of cardiac disease. This entails the achievement of optimal volume status, combined with the appropriate use of antihypertensive agents such as calcium channel blockers, beta-blockers, angiotensin-converting enzyme (ACE) inhibitors, vasodilators, alpha-blockers, and central sympatholytic drugs. In ESRD patients, specific dialysis-related complications such as intradialytic hypotension and pericardial effusion may have additional effects on cardiac function and require attention. The choice of dialysate composition and membrane may influence clinical outcomes with specific effects on cardiac performance.

Continuous registration of blood oxygen tension during hemodialysis

Continuous monitoring of blood oxygen tension was carried out during 45 hemodialyses sessions. The oxygen tension curves displayed different appearances. Some curves presented with a periodicity like sinus curves (type 1), while others (type 2) fluctuated with other appearances, or displayed straight lines. Blood pressure variations during dialysis were significantly greater during dialyses displaying type 1 curves than during those displaying type 2 curves (p values < 0.02). Due to a greater descend of S-urea during dialyses with type 1 curves than during dialyses with type 2 curves (p < 0.05) it is hypothesized, that the intensity of the dialysis treatment could be of some importance to the observed fluctuations in oxygen tension.

Oxygen status during haemodialysis. The Cord-Group

Hypoxia during haemodialysis, mainly acetate, has been reported several times. In our study we have monitored oxygen status during 258 bicarbonate haemodialyses. A significant drop below 80 mmHg in mean oxygen tension occurred. Mean oxygen saturation reflected this drop but did not reach levels below 90%. The mean oxygen concentration was on the whole critical low, though slightly increasing during each haemodialysis session due to ultrafiltration. It is concluded that both hypoxia and hypoxaemia do occur during bicarbonate haemodialysis. To a group of patients generally having limited cardiac reserves, a poor oxygen status is a potentially serious complication to haemodialysis. Monitoring oxygen status is thus advisable.

Biocompatibility of membranes used in the treatment of renal failure

Haemodialysis membranes with a wide range of solute and hydraulic permeabilities are used clinically. Such membranes are manufactured from either cellulose or synthetic co-polymers and their biocompatibility is commonly characterized by the complement activation and white cell changes observed during their use. The cellobiosic unit may be modified by the partial or total replacement of the hydroxyl groups by diethylaminoethyl (Hemophan), acetate (cellulose acetate), triacetate (cellulose triacetate) or 2,5-acetate (Diaphan). We have undertaken a prospective study in which such renal membranes have been studied in terms of the complement activation and neutropenia produced with the aim of investigating the relationship between modification of the cellobiosic unit and the magnitude of neutropenia and complement activation, and the extent to which membrane base material influences these parameters, by comparing the changes observed in modified cellulose membranes with that for a synthetic membrane (polysulphone). Our findings show that, while the degree of substitution varies between < 1% and total substitution, there is no correlation between the numbers of hydroxyl groups replaced and alteration of complement activation and neutropenia. However, by modification of the cellobiosic unit it is possible to produce a membrane whose biocompatibility is similar to that of a membrane manufactured from a synthetic co-polymer such as polysulphone.

Role of bioincompatibility in dialysis morbidity and mortality

This review examines the mechanisms by which bioincompatibility in dialysis systems may have an effect on morbidity and mortality in the dialysis population. Direct toxic effects of membrane materials and various chemical substances have been well demonstrated in the chronic dialysis population. Activation of the complement cascade and stimulation of cytokine production may have autocrine effects on leukocyte function with sequelae such as enhanced rates of infection and the development of B2-microglobulin amyloidosis. The variable effect of different membrane materials on each of these effector systems is examined. Bioincompatibility may effect the incidence of infection, malignancy, cardiopulmonary disease, and malnutrition as well as induce novel disease processes. All these confounding variables must be considered when evaluating the effect of dialysis on mortality and morbidity.

Sleep apnea syndrome and end-stage renal disease. Cure after renal transplantation

We report two patients undergoing maintenance hemodialysis who presented with sleep apnea syndrome (SAS). The first patient is a 36-year-old man with a terminal Berger's glomerulopathy and associated obstructive sleep apnea syndrome (OSAS) (apnea-hypopnea index [AHI] = 80). He was receiving home hemodialysis and was treated by nasal continuous positive airway pressure (CPAP). After successful renal transplantation, his symptoms completely disappeared, and control polysomnography greatly improved (AHI = 9). The second patient had hypokalemic nephropathy with severe, uncontrolled hypertension and hypertensive myocardiopathy. He was receiving home dialysis and showed a central sleep apnea syndrome with an AHI of 51. He also was successfully treated by nasal CPAP. After renal transplantation, his sleep improved, insomnia disappeared, and polysomnography showed great improvement (AHI = 5). We discuss the role of periodic breathing related to end-stage renal disease associated metabolic abnormalities, as a pathogenetic factor of these SASs. Respiratory correction of chronic metabolic acidosis, "uremic toxins," "middle molecules," and hemodialysis are all evoked as etiologic factors and their own roles are discussed.

Continuous blood gas monitoring in haemodialysis using an electrode inserted in the extracorporeal dialysis circulation

The aim of this study was to evaluate the usefulness of continuous gas monitoring during haemodialysis using an oxygen- and carbon dioxide-sensitive electrode inserted in the 'arterial' line from the arteriovenous fistula. We investigated 15 patients with chronic renal failure, treated with regular bicarbonate haemodialysis. The PO2 and PCO2 were measured by the new method and compared to conventional blood gas analysis of the 'arterial' blood from the dialysis circulation. We found the in situ gas sensors were stable and there was no systematic difference between gas tensions measured by in situ and conventional blood gas measurements. The standard deviations of the difference between paired measurements for PO2 and PCO2 were 1.3 and 0.4 kPa, respectively. Therefore, the method can be used to detect major changes in blood gas tensions during haemodialysis.

Pulmonary gas exchange during dialysis in patients with obstructive lung disease

Hypoxemia occurs during routine hemodialysis and may contribute to morbidity, but its cause is not well understood. We reasoned that patients with COPD would be more vulnerable to abnormalities in gas exchange with dialysis. Thus, to investigate the cause of dialysis-related hypoxemia, we measured gas exchange in a group of stable dialysis patients with normal pulmonary function (n = 6) and a group of dialysis patients with COPD (n = 6). Measurements were made predialysis, at 1 h, and postdialysis with both acetate and bicarbonate dialysates. Acetate dialysis decreased PaO2 in normal and COPD patients at 1 h and postdialysis. Acetate-induced hypoxemia was associated with reduced respiratory CO2 excretion and hypoventilation but PaCO2 did not change. This decrease in CO2 excretion resulted from CO2 fixation during acetate metabolism and modest CO2 loss across the dialyzer. Hypoxemia occurred only postdialysis with bicarbonate dialysate in normal and COPD patients. An increased P(A-a)O2 occurred postdialysis with both dialysates, and was most consistently observed in the COPD patients. In summary, at least two mechanisms contribute to dialysis hypoxemia. With acetate dialysate, alveolar hypoventilation from CO2 unloading occurs at 1 h and postdialysis due to acetate metabolism. However, abnormalities in ventilation/perfusion contribute to postdialysis hypoxemia observed with both dialysates. In addition, the decrement in PaO2 associated with dialysis is similar in normal and COPD patients, although preexisting COPD makes postdialysis changes more apparent.