Version May 2024
INTRODUCTION — The dialysis disequilibrium syndrome (DDS) is characterized by a range of neurologic symptoms that affect patients on hemodialysis, particularly when they are first started on dialysis [1,2]. However, it is also seen among patients who have missed multiple consecutive dialysis treatments. This may be due to nonadherence or inability to receive regular dialysis treatments because of economic hardships, such as among undocumented immigrants in the United States. Symptoms of DDS are thought to be primarily due to cerebral edema. DDS has rarely been described in patients treated with continuous kidney replacement therapies (CKRT) [3,4].
This topic reviews the pathogenesis, clinical manifestations, diagnosis, prevention, and treatment of DDS. Other complications of dialysis are covered at length elsewhere:
●(See "Acute complications during hemodialysis".)
●(See "Seizures in patients undergoing hemodialysis".)
EPIDEMIOLOGY AND RISK FACTORS — The incidence of dialysis disequilibrium syndrome (DDS) is poorly defined and varies by patient population. The overall incidence and severity have declined, in part due to preventative strategies and earlier initiation of dialysis [2,5-7]. However, the incidence may also be underreported due to diminished severity and the nonspecific nature of the symptoms [8].
Risk factors for DDS include the following [2,5,6,8,9]:
●First hemodialysis treatment
●Markedly elevated blood urea nitrogen (BUN) concentration prior to a dialysis session (eg, >175 mg/dL or 60 mmol/L)
●Extremes of age
●Preexisting neurologic diseases (head trauma, stroke, seizure disorder)
●Concomitant presence of other conditions that could be associated with cerebral edema (such as hyponatremia, hepatic encephalopathy, or hypertensive emergency)
●Concomitant presence of another condition associated with increased permeability of the blood-brain barrier (such as sepsis, vasculitis, thrombotic thrombocytopenic purpura, hemolytic uremic syndrome, encephalitis, or meningitis)
Some, but not all, studies suggest that the use of continuous kidney replacement therapies (CKRT) or peritoneal dialysis may be associated with a lower risk of DDS [2,3,9-13].
PATHOGENESIS — The symptoms of dialysis disequilibrium syndrome (DDS) are caused by cerebral edema [5], but the mechanism for development of cerebral edema is unclear. Several theories have been proposed.
●Reverse osmotic shift due to urea – Hemodialysis rapidly removes small solutes such as urea. The rapid decline in the blood urea content in patients who have marked azotemia acutely and significantly lowers plasma osmolality, thereby creating a transient osmotic gradient between plasma and brain cells. This gradient leads to water shift into neurons that produces cerebral edema.
Urea is generally considered an "ineffective" osmole because of its ability to permeate cell membranes. However, equilibration of urea across cell membranes may take several hours to reach completion. In the setting of dialysis, where urea is swiftly moved out of the circulation, its continued presence in tissues including brain cells may exert an osmotic force, drawing water into the cells and producing cerebral edema. This force is further enhanced by an adaptive increase in the water channels and decrease in the urea channels in response to uremia [14].
The pathogenic role of rapid urea clearance was demonstrated in a series of rat experiments [15-17]. Uremic rats developed cerebral edema when dialyzed against a bath lacking urea, but not when they were dialyzed against a bath with an isosmotic concentration of urea. Another study included 10 patients with blood urea nitrogen (BUN) concentrations ranging from 210 to 460 mg/dL who underwent low-efficiency dialysis and serial measurement of blood and cerebrospinal fluid (CSF) urea at various intervals before and after dialysis [18]. Before dialysis, CSF urea was lower than blood urea by 10 to 30 mg/dL. After dialysis, CSF urea was higher than blood urea by 30 to 160 mg/dL. This change in the ratio of CSF to blood urea corresponded to an increase in CSF osmolality by 7.6 mOsm/kg, thereby increasing intracranial pressure.
●Reverse osmotic shift due to other osmoles – Other theories proposed to explain development of cerebral edema among patients who initiate dialysis include the fall in the intracellular pH of brain cells [2,19], increase in brain organic osmolytes [15,16,18,20], and increase in carbon dioxide (CO2) retention after dialysis with a higher bicarbonate dialysate [21]. The fall in intracellular pH can cause sodium and potassium that are bound to proteins to dissociate, thereby rendering them osmotically active.
CLINICAL MANIFESTATIONS — Dialysis disequilibrium syndrome (DDS) is a cluster of symptoms that occurs soon after initiation of kidney replacement therapy [12,18]. Patients at high risk for DDS include patients new to dialysis, those who have an extremely high blood urea nitrogen (BUN), or those who have other active neurologic conditions at the time of dialysis.
Patients report headache, nausea, blurred vision, and restlessness that can progress to somnolence, confusion, disorientation [22], or mania. Mild symptoms are usually self-limited in most patients. However, severe manifestations can include seizures, stupor, coma, and death [22].
While DDS is generally described as a symptom cluster occurring at the beginning of a dialysis session, some consider symptoms occurring towards the end of dialysis (eg, muscle cramps, dizziness) to be part of the syndrome [2,19,23-26].
DIAGNOSIS — Dialysis disequilibrium syndrome (DDS) is a clinical diagnosis, and there is no specific diagnostic test. The diagnosis is suspected in patients who present with typical symptoms during their first dialysis treatment, or while resuming dialysis after a period of nonadherence to dialysis treatments. (See 'Clinical manifestations' above.)
Patients who have typical symptoms that are mild and self-limited (resolving over a period of hours) do not require an extensive evaluation to exclude other possible diagnoses.
In other patients, the diagnosis of DDS is one of exclusion. Other disorders that must be excluded are conditions that cause altered mental status, such as uremia itself, subdural hematoma, cerebral infarction, intracerebral hemorrhage, meningitis, metabolic disturbances (hyponatremia, hypoglycemia), posterior reversible encephalopathy syndrome, and drug-induced encephalopathy [19].
The extent to which each of these conditions is evaluated depends upon the presenting symptoms. Patients who present with an altered mental status during the course of dialysis treatment should have serum electrolytes, calcium, and glucose determined to exclude hypo/hyperglycemia, hypo/hypernatremia, or hypo/hypercalcemia. Infections should be excluded with physical examination, chest radiograph, and blood and urine cultures. Neuroimaging is usually required to exclude stroke, an intracranial hemorrhage, or other intracranial processes.
Patients who develop seizures, stupor, or coma should undergo an evaluation for other potential causes of these findings. (See "Diagnosis of delirium and confusional states" and "Psychiatric illness in adults receiving maintenance dialysis", section on 'Delirium' and "Seizures in patients undergoing hemodialysis", section on 'New-onset seizure'.)
PREVENTION — The two main approaches to prevent dialysis disequilibrium syndrome (DDS) include limiting the clearance of blood urea nitrogen (BUN) by lowering the dialysis blood flow and shortening the dialysis session length, and modeling the dialysis sodium concentration in a way that the osmotic shift resulting from urea clearance is significantly mitigated.
Patients being newly initiated on hemodialysis — Limiting removal of urea by dialysis can prevent large osmotic shifts that are implicated in the pathogenesis of DDS. (See 'Pathogenesis' above.)
We use the following approach:
●We typically initiate hemodialysis with a two-hour session using a blood flow of 150 to 250 mL/min and a dialysate flow that is two times the blood flow rate.
●We dialyze patients for their second and third session on consecutive days following their first session (unless the consecutive session falls on a Sunday). Among patients who did not experience symptoms and signs of DDS during the first dialysis session, we increase the blood flow (and correspondingly the dialysate flow) by 50 mL/min and dialysis time by 30 minutes for the second treatment. If symptoms and signs of DDS occurred during the first session, we similarly increase the intensity but also perform sodium modeling. (See 'Patients with recurrent nonadherence to hemodialysis' below.)
●The prescription for the third treatment should match that which the patient will be receiving as an outpatient. Thus, for the third treatment, we increase the blood flow to a maximum of 400 mL/min, dialysate flow to 800 mL/min, and dialysis time to a maximum of four hours.
●Among patients who have an extremely elevated BUN, such as among patients with a BUN >100 mg/dL, or neurologic symptoms such as myoclonus, disorientation, or somnolence, dialysis should be initiated as an inpatient.
Patients with recurrent nonadherence to hemodialysis — We use sodium modeling to mitigate osmotic shifts among patients who frequently miss dialysis. It is typically not feasible to admit such patients to the hospital each time for slow re-initiation of dialysis. As an example, undocumented immigrants in the United States are often uninsured or underinsured and, therefore, miss dialysis and cannot be repeatedly hospitalized for re-initiation. In these cases, we perform sodium modeling for two dialysis sessions to prevent DDS before resuming the usual prescription.
Our approach to modeling sodium depends upon whether or not the patient is being dialyzed using a machine that is capable of automatically modeling sodium and upon the patient's predialysis serum sodium [22]:
●If the dialysis machine has sodium modeling capability, then we use either linear or exponential modeling profiles. The initial and final dialysate sodium concentrations vary depending upon the expected urea clearance during the treatment. As an example, if 50 percent urea clearance is expected, we set the initial dialysate sodium to be 15 mEq/L higher than the patient's predialysis serum sodium and the final dialysate sodium to be 5 mEq/L higher than the patient's predialysis serum sodium (an average of 10 mEq/L higher). If more robust urea clearance is anticipated, we set higher values; if less urea clearance is expected, we set lower values.
●If the dialysis machine does not have sodium modeling capability, then (if 50 percent urea clearance is expected) we dialyze using a dialysate sodium that is 10 mEq/L higher than the patient's predialysis serum sodium. As above, we use a higher sodium bath if more robust urea clearance is anticipated, and a lower sodium bath if less urea clearance is expected.
If there is capability to model sodium, either linearly or exponentially, then we prefer to use that feature to avoid hypertonicity at the end of the session.
Sodium modeling is essential because the degree of osmotic shift may be variable among individuals and depends upon the dialyzer, time on dialysis, sex, and body size of the patient. Our practice for sodium modeling is largely supported by our clinical experience in addition to a limited evidence base. In one trial, 17 patients were randomly assigned to receive dialysis with higher dialysate sodium compared with standard care [22]. Approximately one-half of the patients in each group received dialysis for the first time as part of the study. All patients underwent serial electroencephalograms (EEG) prior to, during, and hourly up to three hours after dialysis, which were reviewed by a clinician blinded to the randomization. DDS was diagnosed if there was onset of one severe (psychosis, convulsion, stupor, or coma), two major (asterixis, myoclonus, somnolence, or disorientation), or three minor (headache, vomiting, drowsiness, restlessness, or muscle cramps) symptoms. Delivered dialysate had a mean sodium of 133 mEq/L in the control group and 149 mEq/L in the intervention group. The mean increase in sodium was 8 mEq/L, with a reported maximum of 154 mEq/L in the intervention group. Nine of 13 patients in the control group and no patients in the intervention group manifested DDS. EEG abnormalities were present in 10 of 13 patients in the control and two of nine patients in the intervention group. Lack of blinding of the outcome adjudicators was a limitation of this study and may explain the high incidence of DDS in the control group.
Although sodium modeling helps limit osmotic disequilibrium, patients may experience increased thirst, higher intradialytic weight gain, and higher blood pressures. Thus, these variables must be monitored closely, and additional sessions of isolated ultrafiltration must be planned among patients who are or become volume overloaded as a result of sodium modeling.
Additional measures in patients with carbon dioxide retention — In patients with carbon dioxide (CO2) retention, such as those with chronic obstructive pulmonary disease (COPD), special attention should be given to the prescription of dialysate bicarbonate concentration. If chronic CO2 retention is a concern, then we use a dialysis bicarbonate concentration of 30 mEq/L rather than the standard 35 mEq/L. This is because CO2 is a potent cerebral vasodilator and can provoke an increase in intracranial pressure [21].
TREATMENT — We treat dialysis disequilibrium syndrome (DDS) by modifying the dialysis prescription, regardless of the severity of the symptoms. Residual mild symptoms that are present after adjustment of the prescription should be treated symptomatically. With appropriate management, most patients recover within 24 hours [27]. If improvement is not seen, other etiologies for the alteration in mental status should be considered. (See "Delirium and acute confusional states: Prevention, treatment, and prognosis".)
Initial management — Among patients who develop DDS, we initially review the delivered dialysis prescription and quickly initiate sodium modeling (either by engaging this feature on the dialysis machine or by changing the dialysate sodium bath). We suggest not taking the patient off dialysis. In our experience, the clinical improvement in response to this intervention can be rapid (usually within 30 minutes), such that taking a patient off dialysis may not be necessary.
Among patients with any degree of DDS symptoms, ranging from nausea to coma, we use sodium modeling to mitigate symptoms. The method for sodium modeling is the same as described for prevention of DDS. (See 'Patients with recurrent nonadherence to hemodialysis' above.)
If clinical improvement is not noted within 30 minutes, it is reasonable to take the patient off dialysis and investigate other causes of their symptoms.
●(See "Evaluation and management of the first seizure in adults".)
●(See "Acute complications during hemodialysis".)
●(See "Diagnosis of delirium and confusional states".)
●(See "Stupor and coma in adults".)
Patients who fail to respond to initial measures — Among patients with persistent severe DDS (such as seizures, encephalopathy, or coma) despite the use of sodium modeling, a trial of hypertonic saline or mannitol is reasonable, where available. We use either 5 mL of 23 percent saline or 12.5 g of mannitol to rapidly raise the serum osmolality and to prevent further osmotic shifts. This practice is largely based upon anecdotal clinical experience as data in support remain lacking.
Once hypertonic saline or mannitol has been administered, we stop dialysis and plan for daily, short low-efficiency dialysis sessions, similar to patients being newly initiated on dialysis. (See 'Patients being newly initiated on hemodialysis' above.)
In addition, patients who continue to experience mild residual symptoms of DDS (nausea, vomiting, restlessness) after modification of the prescription and use of sodium modeling can be managed symptomatically [27].
Acute management of a patient with severe symptoms of DDS, such as seizure, delirium, and coma, is discussed at length elsewhere. In such settings, improvement should occur within 24 hours [27]. The patient should also undergo evaluation for other causes of their symptoms.
●(See "Acute complications during hemodialysis".)
●(See "Evaluation and management of the first seizure in adults".)
●(See "Diagnosis of delirium and confusional states".)
●(See "Stupor and coma in adults".)
SOCIETY GUIDELINE LINKS — Links to society and government-sponsored guidelines from selected countries and regions around the world are provided separately. (See "Society guideline links: Dialysis".)
SUMMARY AND RECOMMENDATIONS
●Overview – Dialysis disequilibrium syndrome (DDS) is characterized by a range of neurologic symptoms that affect patients on hemodialysis when they are first started on dialysis or when patients have missed multiple consecutive dialysis treatments. DDS in patients starting on continuous kidney replacement therapy (CKRT) has also been described. (See 'Introduction' above.)
●Risk factors – New patients being initiated on intermittent hemodialysis are at greatest risk for DDS, particularly if the blood urea nitrogen (BUN) is markedly elevated (eg, >175 mg/dL or 60 mmol/L). Other predisposing factors include extremes of age, preexisting neurologic diseases (eg, head trauma, seizure disorder), conditions associated with an increased permeability of the blood-brain barrier (eg, encephalitis), and concomitant presence of other conditions that are associated with cerebral edema (eg, hepatic encephalopathy). (See 'Epidemiology and risk factors' above.)
●Pathogenesis – The symptoms of DDS are caused by cerebral edema, but the mechanism for development of cerebral edema is unclear. The prevailing theory is that rapid clearance of urea and other osmoles by hemodialysis results in a rapid fall in the plasma osmolality, which leads to movement of water into the neurons. (See 'Pathogenesis' above.)
●Clinical manifestations – Clinical manifestations of DDS include headache, nausea, blurred vision, and restlessness that can progress to somnolence, confusion, disorientation, or mania. Mild symptoms are usually self-limited in most patients. However, severe manifestations can include seizures, stupor, coma, and death. (See 'Clinical manifestations' above.)
●Diagnosis – DDS is a clinical diagnosis, and there is no specific diagnostic test. The diagnosis is suspected in patients who present with typical symptoms during their first dialysis treatment, or while resuming dialysis after a period of nonadherence. In patients with symptoms that are persistent or severe, the diagnosis of DDS is one of exclusion. Other disorders that must be excluded are conditions that cause altered mental status, such as uremia, subdural hematoma, cerebral infarction, intracerebral hemorrhage, meningitis, metabolic disturbances (hyponatremia, hypoglycemia), and drug-induced encephalopathy. (See 'Diagnosis' above.)
●Prevention – The two main approaches to prevent DDS include limiting the clearance of BUN by lowering the dialysis blood flow and shortening the dialysis session length, and modeling the dialysis sodium concentration in a way that the osmotic shift resulting from urea clearance is significantly mitigated. Our choice of approach depends upon the patient group and is as follows (see 'Prevention' above):
•Among patients who are newly initiated on hemodialysis, we start with a short session length (ie, a two-hour session) using a low blood flow (ie, 150 to 250 mL/min) and gradually increase the treatment time and blood flows with subsequent sessions. Among patients who have an extremely elevated BUN (eg, BUN >100 mg/dL), or neurologic symptoms such as myoclonus, disorientation, or somnolence, dialysis should be initiated as an inpatient. (See 'Patients being newly initiated on hemodialysis' above.)
•Among patients with recurrent nonadherence to hemodialysis for whom repeated hospitalization for slow re-initiation is not feasible, we perform sodium modeling for two sessions. Our method for sodium modeling is outlined above. (See 'Patients with recurrent nonadherence to hemodialysis' above.)
•Among patients with chronic carbon dioxide (CO2) retention, we use a dialysis bicarbonate concentration of 30 mEq/L rather than the standard 35 mEq/L. This is because CO2 is a potent cerebral vasodilator and can provoke an increase in intracranial pressure. (See 'Additional measures in patients with carbon dioxide retention' above.)
●Treatment – We treat DDS by modifying the dialysis prescription, regardless of the severity of the symptoms. With appropriate management, most patients recover within 24 hours. If improvement is not seen, other etiologies for the alteration in mental status should be considered. (See 'Treatment' above.)
•Among patients who develop DDS, we initially review the delivered dialysis prescription and quickly initiate sodium modeling (either by engaging this feature on the dialysis machine or by changing the dialysate sodium bath). We suggest not taking the patient off dialysis. If clinical improvement is not noted within 30 minutes, it is reasonable to take the patient off dialysis and investigate other causes of their symptoms. (See 'Initial management' above.)
•Among patients with persistent severe DDS (such as seizures, encephalopathy, or coma) despite the use of sodium modeling, a trial of 5 mL of 23 percent saline (hypertonic saline) or 12.5 g of mannitol is reasonable, if available. Once hypertonic saline or mannitol has been administered, we stop dialysis and plan for daily, short low-efficiency dialysis sessions, similar to patients being newly initiated on dialysis. (See 'Patients who fail to respond to initial measures' above.)
ACKNOWLEDGMENT — The UpToDate editorial staff acknowledges Lionel U Mailloux, MD, FACP, who contributed to earlier versions of this topic review.
آیا می خواهید مدیلیب را به صفحه اصلی خود اضافه کنید؟
