Management of hypervolemia in patients on peritoneal dialysis

INTRODUCTION — The goal of peritoneal dialysis is to replace as much native kidney function as possible. As such, a major aim is to remove solute and water so that the patient does not accumulate uremic solutes or excess fluid. Volume overload is a significant cause of peritoneal dialysis failure and transfer to hemodialysis, and optimal control of volume status is an important aspect of administering high quality peritoneal dialysis [1].

The volume status of individual patients changes continuously and requires ongoing adjustment of the dialysis regimen. This is usually done at home by the patients, with the assistance of the home dialysis staff, under the supervision of the prescribing clinician. Occasionally, hypervolemia develops despite these adjustments, however, and must be evaluated.

This topic reviews factors that contribute to hypervolemia in patients on peritoneal dialysis. We provide an approach to the evaluation and management of patients on peritoneal dialysis who develop progressive hypervolemia despite dialysis.

The management of patients on peritoneal dialysis with inadequate solute clearance is discussed elsewhere. (See "Inadequate solute clearance in peritoneal dialysis".)

MECHANISM OF FLUID REMOVAL — Peritoneal dialysis removes fluid by ultrafiltration. Ultrafiltration is the passage of fluid across a semipermeable membrane in response to a driving pressure (ie, hydrostatic, oncotic, or osmotic pressure).

Water moves from the capillary network perfusing the peritoneal cavity to the peritoneal dialysate fluid via intercellular and transcellular "pores" in the peritoneal membrane [2-4]. Whereas both water and solute are able to cross the intercellular pores, the transcellular pore (also called the aquaporin channel) allows only water transport [5].

Fluid crosses these pores in response to an osmotic gradient that is generated by dextrose (or occasionally the dextrose polymer, icodextrin), which is present in peritoneal dialysate. Fluid movement is also influenced by oncotic gradients generated by albumin and other proteins in blood, by the dwell time, and by the transport characteristics intrinsic to the peritoneal membrane.

Ultrafiltration is typically increased by increasing the concentration of peritoneal dialysate dextrose. Peritoneal dialysis solutions are available with variable amounts of dextrose (ie, 1.5, 2.5, and 4.25 percent). Patients are instructed to increase the dialysate dextrose if their body weight goes above a specific target weight, reflecting hypervolemia, or if signs of hypervolemia (ie, edema, hypertension) develop (see 'Monitoring' below). Net ultrafiltration per dwell is also influenced by dwell time and transport type.

HYPERVOLEMIA

Epidemiology and causes — Hypervolemia is common in patients on peritoneal dialysis [6-8]. In a study of 639 individuals who had been on peritoneal dialysis for over two years, fluid overload was present in more than 70 percent of patients [9].

Causes may be unrelated to the dialysis procedure (such as loss of residual kidney function or excessive sodium and water intake) or related to the dialysis procedure. In addition, use of a new medication, a new cardiovascular event, or worsening of underlying heart disease may lead to signs of hypervolemia, such as pedal or pulmonary edema.

Unrelated to dialysis

●Loss of residual kidney function ─ Patients who initiate peritoneal dialysis usually have some residual kidney function that contributes to fluid clearance. Residual kidney function tends to decline over time in most patients. This reduces net fluid removal unless the dialysis prescription is adjusted to provide more ultrafiltration.

●Excessive dietary sodium and water intake ─ We generally restrict sodium to <2 grams/day. We do not restrict fluids, at least initially, if the patient has residual kidney function. However, many individuals will start to require fluid restriction as residual kidney function declines. Excessive sodium and fluid intake in such patients will cause hypervolemia. (See 'Additional preventive measures' below.)

Related to dialysis

●Lack of adherence – Patients occasionally skip exchanges, shorten the daytime dwell, or lengthen the nighttime dwell. Depending on the rate of solute transport, shortening the daytime dwell may not allow sufficient time for ultrafiltration. Lengthening the nighttime dwell can cause reabsorption of ultrafiltered fluid back into the blood or lymphatic system. Some patients may use the incorrect solution during the overnight dwells or fail to adjust the dialysate dextrose concentration based on body weight (which results in an insufficient osmotic stimulus for the needed ultrafiltration). Some patients shorten the drain time and do not wait for complete drain before instilling new fluid. This results in the accumulation of fluid and eventual reabsorption of fluid into blood and lymphatics or dilutes the dextrose (osmotic agent) concentration of the subsequent dwell.

●Incorrect dialysis prescription for transport type – The dialysis prescription can be optimized according to specific membrane transport characteristics for each individual patient. These transport characteristics are defined by a peritoneal equilibration test (PET) that is performed shortly after the start of dialysis. If the transport characteristics change, the prescription may need to be adjusted accordingly. As an example, "fast or high transporters" have peritoneal membranes that allow more rapid movement (absorption) of dextrose across the membrane, thereby rapidly dissipating the osmotic gradient. Since dextrose provides a major osmotic stimulus for water movement, such patients will reach their peak ultrafiltration volumes early in a dwell (approximately two to four hours depending on dialysate dextrose concentration). After this time, ultrafiltration volumes start to decline as reabsorption occurs. Thus, “fast transporters” are often better served by a shorter dwell time. By contrast, "slow or low transporters" have peritoneal membranes that exhibit slower transport of solute (and, therefore, slower absorption of dextrose) so longer dwell times can be tolerated. In these patients, maximal ultrafiltration occurs at four to six hours, depending on dialysate dextrose concentration. The relationship between membrane transport characteristics and ultrafiltration is described in detail elsewhere (figure 1). (See "Peritoneal equilibration test".)

●Hypoalbuminemia and hyperglycemia – Hypoalbuminemia from chronic inflammation, poor nutrition, or liver disease may decrease the oncotic pressure gradient that drives water from the interstitial space into the intravascular space, which can then be removed from the blood into dialysate. It is difficult to remove fluid in such patients and, therefore, hypervolemia is common. In our experience, hypervolemia from hypoalbuminemia usually does not occur unless the serum albumin concentration is less than 2.8 g/dL.

Chronic hyperglycemia may reduce the osmotic pressure gradient between blood and dialysate that drives ultrafiltration, leading to retention of excess fluid and hypervolemia.

●Mechanical problems causing collection of dialysate – Anything that causes a chronic (and usually unrecognized) collection of dialysis fluid may result in hypervolemia. Catheter malposition, multiple adhesions, or problems such as retroperitoneal fluid leaks may allow normal infusion of dialysis fluid but insufficient draining of dialysate plus ultrafiltrate. This causes a fluid collection that gets absorbed over time into capillaries and lymphatics. A common example is a catheter leak into retroperitoneal or abdominal wall space.

Incomplete draining of peritoneal fluid may also result in a large residual volume in the peritoneal cavity that dilutes the dextrose concentration of the next dwell. This decreases the osmotic stimulus for ultrafiltration. As an example, the 4.25 percent dextrose solution that is instilled ends up with less than 4.25 percent dextrose, and the 7.5 percent icodextrin ends up containing less than 7.5 percent icodextrin. Depending on the amount of dilution, this could affect the ability to remove fluid [10].

●Ultrafiltration failure – With true ultrafiltration failure (or "membrane failure"), there is insufficient fluid transport across the peritoneal membrane despite optimal conditions (ie, adherence, correct prescription, adequate albumin, etc).

Ultrafiltration failure can occur at any stage of peritoneal dialysis [11]. The reported incidence is between 10 and 40 percent [11,12]. However, this is probably an overestimate since apparent ultrafiltration failure is often just a mismatch of dwell time to transport type that can be addressed by adjusting the dialysis prescription. True ultrafiltration failure often results in transfer to hemodialysis.

The major risk factor for ultrafiltration failure is repeated episodes of peritonitis. Other reported risk factors include peritoneal dialysis of greater than two years duration (even in the absence of peritonitis), increased exposure to high-concentration dialysate glucose, and possibly diabetes mellitus and the use of beta blockers [11,13-15].

Membrane-related true ultrafiltration failure has traditionally been categorized into three types. A fourth type of ultrafiltration failure (type 4) has clinical characteristics of membrane failure, but the membrane transport function is intact. The different types of ultrafiltration failure are as follows:

•Type 1 – There is very rapid membrane solute transport (ie, "rapid or high transporters"), resulting in rapid equilibration and dissipation of the osmotic gradient. Type 1 is the most common type of ultrafiltration failure [12] and may be transient during an acute episode of peritonitis [16,17] or sustained following repeated episodes [18].

•Type 2 – There is decreased function of aquaporins, resulting in isolated decrease in water transport. This condition is rare.

•Type 3 – There is loss of effective peritoneal membrane surface area due to sclerosis or adhesions, resulting in decreased permeability to both solute and fluid.

•Type 4 – There is an increase in the rate of absorption of fluids from the peritoneal cavity via lymphatic vessels and into local tissue. This presents in a similar fashion to true membrane failure although peritoneal membrane function is intact [19]. In such cases, even with adequate ultrafiltration of fluid from capillaries into dialysate, the ultrafiltered fluid is absorbed instead of draining via the peritoneal catheter to the outside. Normally, the rate of reabsorption is approximately 1 mL/min. Hypervolemia may result when absorption rates are greater than 3 mL/min.

Increased absorption may be related to excessive intraperitoneal pressure, which occasionally occurs when large instilled volumes are used in small patients; reducing instilled volume often resolves the problem in such cases. It appears to be unrelated to the duration of peritoneal dialysis use [20].

Prevention — The most important measure to prevent hypervolemia is to establish and maintain an accurate target weight. It is also important to monitor urine volume and the volume of fluid removed with dialysis so that the dialysis prescription may be modified before significant hypervolemia develops. All measures are discussed below.

Establish target weight — The target weight is defined as the weight at which the patient does not have hypertension or other evidence of fluid overload (eg, pulmonary or pedal edema, ascites) and does not have evidence of volume depletion (orthostatic symptoms, tachycardia). The initial target weight is established by trial and error during peritoneal dialysis training.

An accurate target weight is important because, unlike hemodialysis, peritoneal dialysis is a home therapy that is carried out by patients who are typically examined by their clinician only once per month. Thus, the clinical judgment that determines targets of fluid removal is rendered by the patient. At most centers, patients are trained to adjust the dextrose concentration of the peritoneal dialysis fluid based on their current weight (in comparison to the defined target weight), blood pressure, or physical examination evidence of lower-extremity edema.

The target weight tends to change over time. Patients may gain or lose actual body mass (eg, muscle mass, fat mass) while on dialysis. Underestimates or overestimates in target weight will result in volume depletion or hypervolemia, respectively. The target weight often must be increased gradually for several months after starting dialysis, when resolution of uremic-associated anorexia can lead to increases in lean body mass and adipose tissue.

Monitoring — Patients are generally evaluated at least monthly:

●We routinely re-evaluate target weight based on clinical exam, dietary history, and by any changes in cardiovascular status. If a patient develops pedal and pulmonary edema at the prescribed target weight, they have likely lost body mass, and the target weight needs to be set lower to better approximate euvolemia.

●We review the ultrafiltration volumes achieved and the percent dextrose used with each exchange in order to monitor the ultrafiltration capability of the membrane and the total fluid volume removed daily with dialysis.

For patients on automated peritoneal dialysis (APD), the ultrafiltration volumes are available from data recorded by the automated cycler; otherwise, we rely on a patient-kept log.

●We measure a 24-hour urine volume to detect any decreases in urine volume that could contribute to a decrease in net fluid removal. If the measurement is considered to be accurate and a decline is detected, we adjust the dialysis prescription to provide more ultrafiltration. (See "Prescribing peritoneal dialysis", section on 'Fluid balance'.)

●We review the patient's medication list. A major concern is a new nephrotoxic medication, prescribed by a clinician who is unaware of the importance of residual kidney function in patients on dialysis. Alternatively, some medications such as dihydropyridine calcium channel blockers may cause peripheral edema due to redistribution of fluid from the intravascular space to the interstitium. In such cases, the patient may have peripheral edema, but if otherwise asymptomatic, no change in dialysis prescription may be needed.

●Most clinicians perform a baseline PET shortly after the initiation of dialysis and repeat the PET only if required for evaluation of decreased ultrafiltration (see 'Peritoneal equilibration test' below). This is our approach and is consistent with the Kidney Disease Outcomes Quality Initiative (KDOQI) guidelines [21].

Additional preventive measures — The following measures may also help to prevent hypervolemia:

●We use diuretics in all patients with residual kidney function, even if they do not have evidence of hypervolemia. The more urine volume a patient has, the less they need to rely on peritoneal dialysis ultrafiltration. Typically, patients are already on loop diuretics (usually furosemide) when they initiate peritoneal dialysis. We generally continue the same dose at the start of peritoneal dialysis and increase as indicated. Our ceiling dose of furosemide is 160 mg three times a day.

●We try to preserve residual kidney function. The preservation of residual kidney function is associated with better outcomes [22-24]. In a report of nearly 2700 patients, each mL per minute increase in the renal creatinine clearance, but not the peritoneal dialysis clearance, was associated with a 12 percent reduction in the odds ratio for death [22].

Specific measures include avoidance of nephrotoxic agents and volume depletion, rigorous control of blood pressure, and the use of angiotensin-converting enzyme (ACE) inhibitors or angiotensin receptor blockers (ARBs) in the setting of proteinuria [25]. Two randomized trials have shown that the use of ACE inhibitors or ARBs preserves residual kidney function in patients on peritoneal dialysis [26,27]. In both studies, the effect of the ACE inhibitor or ARB was independent of blood pressure control [26,27].

●We provide dietary counseling to all patients. This is important in preventing hypervolemia but is frequently overlooked. We typically restrict sodium to 2 g/day or use a no-added-salt diet [28]. We do not restrict fluid unless the patient has no residual kidney function.

●We encourage adherence in maintaining the peritoneal dialysis regimen.

●We optimize serum glucose control. As noted above, patients with consistently elevated blood glucose lose the osmotic gradient between the blood and peritoneal fluid that drives ultrafiltration. We review HbA1C levels at least twice a year. Management of diabetes in patients on peritoneal dialysis is presented separately. (See "Management of hyperglycemia in patients with type 2 diabetes and advanced chronic kidney disease or end-stage kidney disease".)

●We try to limit the use of 4.25 percent exchanges in such patients since the higher glucose concentration is more likely to be associated with hyperglycemia. Icodextrin may be used instead of 4.25 percent dextrose for one dwell (usually the longest dwell of the day). Icodextrin is slowly absorbed and allows for better ultrafiltration compared with dextrose. What little is absorbed is eventually metabolized to glucose such that the caloric load is approximately equivalent to that of a 2.5 percent dextrose dialysate. (See 'Icodextrin dialysate' below.)

Evaluation — We follow a stepwise and streamlined evaluation for the cause of hypervolemia (algorithm 1). A targeted history and physical, review of dialysis records, a "fill and drain test," and imaging are to identify or exclude the following causes of hypervolemia:

●New or worsening heart disease

●Excessive sodium and water intake

●Inaccurate target weight

●Nonadherence

●Decreased residual kidney function

●Mechanical problem causing fluid accumulation

Clinical history — The patient should be specifically asked about the following:

●Symptoms of heart disease – New or worsening heart diseases should be further evaluated as necessary with an electrocardiogram (ECG), echocardiographic, or radiographic studies. Changes in cardiac output could be due to a change in a known arrhythmia or onset of a new arrhythmia such as atrial fibrillation. (See "Heart failure: Clinical manifestations and diagnosis in adults".)

●Dietary history – The patient should be asked about loss of appetite or decreased oral intake. Any loss of body mass could cause the target weight to be artificially high and inaccurate. Readjusting the target weight down effectively treats hypervolemia in such cases.

The patient should also be asked about sodium and water intake, particularly if they do not have residual kidney function. (See 'Sodium and fluid restriction' below.)

●New medication use – The recent addition of nephrotoxic medications may have decreased residual kidney function.

●Adherence with dialysis prescription – Patients should be asked about adherence, specifically including the duration of the overnight dwell, duration of the drain for each exchange, and whether the dialysate dextrose concentration is being adjusted based on body weight, blood pressure, or evidence of hypervolemia such as pedal edema, as instructed. If an automated device is used for exchanges, then clinicians can review the dialysis-related data from the machine.

Physical examination — We use the physical exam to identify mechanical problems, including hernias or catheter leaks, that could cause the accumulation of fluid.

We examine the pericatheter, genital, inguinal, and femoral areas for hernias. We look for evidence of retained fluid that may be unilateral or localized to certain areas, particularly the pleural space, abdominal wall, inguinal areas, and/or genitalia. Any evidence of fluid accumulation should be further evaluated with imaging. (See 'Radiographic studies' below.)

We examine the exit site for the presence of fluid that is dextrose positive on dipstick, which suggests a peritoneal leak (since peritoneal fluid contains dextrose). Not all peritoneal leaks result in leakage of fluid around the exit site, however, and a retroperitoneal or abdominal wall leak could still be present in the absence of dextrose-positive fluid at the exit site.

Dialysis data and urine volume — As for all routine monthly visits, we review manual logs or treatment data from automated cyclers. It is particularly important to determine the drain volume following the long daytime dwell in patients on the automated cycler since reabsorption of fluid is common. We also note the number and type of alarms with the automated cycler since they indicate changes in fluid volume or flow rate that could indicate incomplete draining.

We measure the urine volume to make sure it has not substantially decreased. We review the dialysis supplies to ensure adherence.

Fill and drain test — In patients who do not have an obvious explanation for their hypervolemia after history and physical, we perform a "fill and drain" test under direct visualization to exclude a mechanical problem, such as a catheter leak, malpositioning, or obstruction to outflow that could cause the accumulation of fluid inside the body. Even very rapid transporters will not lose intraperitoneal volume during a quick inflow and outflow test, so diminished outflow in this setting suggests a mechanical problem.

We observe the rate of inflow or outflow since a decreased rate suggests a partial obstruction. We observe for fibrin clots that could cause intermittent obstruction. Positional changes in fluid outflow suggest catheter malpositioning.

The drain volume should be at least equal to the fill volume. A drain volume less than that suggests the accumulation of fluid inside the body.

A detailed discussion of the different mechanical and anatomical abnormalities associated with peritoneal dialysis catheters is found separately. (See "Noninfectious complications of peritoneal dialysis catheters".)

Any evidence of inadequate drainage should be investigated with radiographic studies. (See 'Radiographic studies' below.)

Radiographic studies — We obtain abdominal radiographs in the supine anteroposterior and lateral decubitus positions to ensure optimal catheter positioning. The flat plate may also show constipation, even if denied by the patient. Constipation may cause failure to drain if it causes the tip of the catheter to be located in a "pseudo" pocket in the pelvis created by the rigid sigmoid. The lateral film is done to exclude a rotation or kink in the subcutaneous portion of the catheter.

If the flat plate and lateral films are nonrevealing, either magnetic resonance imaging (MRI) without contrast or a computed tomography (CT) of the abdomen and pelvis with intraperitoneal contrast is performed to detect a catheter leak [29].

MRI is the preferred study since it can distinguish peritoneal fluid accumulation from the surrounding tissue or interstitial fluid even without contrast. (See "Modalities for the diagnosis of abdominal and thoracic cavity defects in patients on peritoneal dialysis".)

We do not use gadolinium among patients on dialysis, since the administration of gadolinium has been associated with the potentially severe syndrome of nephrogenic systemic fibrosis. (See "Nephrogenic systemic fibrosis/nephrogenic fibrosing dermopathy in advanced kidney disease".)

If a CT is used to identify a possible leak, a baseline CT of the abdomen and pelvis with no intraperitoneal fluid or contrast is first obtained. We then add 100 to 200 mL of nonionic contrast to a 2 L dialysate bag, which is instilled in the peritoneum. The patient then ambulates to distribute the fluid (ideally for two hours but for a minimum of 30 minutes). The CT including abdomen and pelvis is repeated.

A catheter flow study (catheterogram) may also be used to identify an omental wrap as a cause of outflow failure. In a catheter flow study, a 40 mL solution containing 20 mL of iodinated contrast and 20 mL of normal saline is rapidly infused into the catheter. Plain radiographs of the abdomen are obtained at 10 seconds, 30 seconds, and two minutes [30].

Peritoneal equilibration test — If we can find no explanation for hypervolemia, and particularly if the dialysis data from patient logs or automated cycler shows persistently low drain volumes despite the use of 4.25 percent dextrose dialysate (suggesting decreased ultrafiltration), we perform a PET (figure 2).

The PET diagnoses ultrafiltration failure and may allow identification of the type. (See "Peritoneal equilibration test".)

Even in the absence of true ultrafiltration failure, by comparison to the baseline PET obtained at the initiation of peritoneal dialysis, the PET may identify changes in solute transport rate that inform modification of the dialysis prescription.

If a PET test is performed as part of the evaluation of hypervolemia, we use 4.25 percent dextrose fluid, rather than a 2.5 percent solution, which may be used for the baseline PET (see "Peritoneal equilibration test", section on 'Elements of the peritoneal equilibration test'). Use of the 4.25 percent dextrose increases the sensitivity of the test for identifying ultrafiltration failure [31,32]. The 4.25 percent PET can be compared with tests performed with either 2.5 or 4.25 percent dextrose. Although there is more ultrafiltration with the 4.25 percent dextrose PET than the 2.5 percent dextrose PET, there is little difference in dialysate/plasma (D/P) ratios for creatinine, which are used for classification of transport type [32].

Laboratory values from the peritoneal fluid are obtained at baseline, at 30 minutes (for dialysate sodium), and at one, two, and four hours [31].

Parameters measured, obtained, or calculated should include the ratio of concentrations in D/P values for creatinine, urea, and sodium and the ratio of dialysate concentrations at specific times (t) to the initial level in the dialysis solution (Dt/Do) values for glucose and drain volumes. Interpretation of the PET is discussed in depth elsewhere. (See "Peritoneal equilibration test", section on 'Diagnosis of causes of inadequate ultrafiltration and solute clearance' and "Peritoneal equilibration test", section on 'Diagnosis of early ultrafiltration failure'.)

The PET provides the following information:

●Diagnoses ultrafiltration failure – An ultrafiltration volume less than 400 mL after a four-hour dwell with 2 L of 4.25 percent dextrose defines ultrafiltration failure. This does not determine the type (ie, cause or underlying pathophysiology) of ultrafiltration failure.

●Characterizes solute transport – Patients are characterized as fast (or high) transporters or slow (or low) transporters using D/P ratios for creatinine. This helps to distinguish the types of ultrafiltration failure.

●Determines whether transcellular aquaporin channels are functional by sodium sieving – The PET provides an indirect assessment of aquaporin activity. If water is moving across aquaporin channels, dialysate sodium decreases during the first 30 to 60 minutes of the dwell. This decrease is called sodium sieving. As aquaporin-mediated ultrafiltration diminishes, sodium moves from plasma to dialysate by diffusion so that the dialysate sodium concentration equilibrates with plasma. Decreased ultrafiltration due to functional impairment of aquaporin activity is reflected by a decrease in sodium sieving [19].

Many centers do not routinely assess sodium sieving in the evaluation of decreased ultrafiltration, since a PET without sodium dialysate measurements usually provides sufficient information. An isolated decrease in water transport due to impaired aquaporin function (ie, Type 2 ultrafiltration failure) is rare. (See 'Related to dialysis' above.)

Once ultrafiltration failure is diagnosed, the type of ultrafiltration failure is distinguished using the D/P creatinine ratio and the absence or presence of sodium sieving:

●Type 1 – In patients with type 1 ultrafiltration failure (ie, extremely rapid solute transport), there is an increase in the D/P creatinine ratio. Sodium sieving may be observed at 30 minutes but is diminished at one hour and may be absent by two hours.

●Type 2 – In patients with type 2 failure (ie, a loss of aquaporin function), the D/P ratio is unchanged, and sodium sieving is decreased at 30 minutes, one, or two hours.

●Type 3 – In patients with type 3 failure (ie, decreased solute transport related to decreased functional surface area), both the D/P ratio and sodium sieving are decreased.

●Type 4 (ie, increased lymphatic and postcapillary absorption) – In patients with increased absorption, the PET diagnosis is ultrafiltration failure, but, since membrane transport function is intact, the D/P values and sodium sieving are normal.

Increased absorption can be detected by adding a marker to dialysate when doing the PET test (labeled albumin). If there is significant absorption, there is rapid uptake of the labeled albumin from the peritoneal cavity. Normally, the rate of absorption is approximately 1 mL/min, with hypervolemia potentially occurring with lymphatic absorption rates greater than 3 mL/min.

However, this test is rarely performed, and the diagnosis is usually made by exclusion.

Treatment — Treatment consists of dietary sodium and fluid restriction, increased dosing of loop diuretics or the addition of combination diuretics, and the use of icodextrin as an osmotic solute during a long dwell [21].

If solute transport characteristics have changed, we may modify the dialysis prescription to match the patient's individual transport type based on the PET.

Sodium and fluid restriction — We restrict sodium intake to <2 grams/day. We restrict fluids to approximately 2 L/day.

Loop diuretics — We treat all patients who have residual kidney function with diuretics. We increase the dose in patients who develop hypervolemia. Furosemide may be given at doses up to 160 mg three times per day in order to obtain higher urine volumes. Combination therapy with furosemide (200 mg orally twice daily) and a thiazide diuretic (eg, metolazone, 5 to 10 mg orally twice daily) may also be effective [33]. Although these approaches increase urine volume, diuretics do not increase urea or creatinine clearances [34].

Icodextrin dialysate — We use icodextrin-containing dialysate for one exchange per day. We use icodextrin-containing dialysate during the overnight exchange in continuous ambulatory peritoneal dialysis [CAPD] and for the long-day dwell in APD.

The use of icodextrin dialysate enhances ultrafiltration, particularly during a long dwell in fast transporters [35-37]. Since the compound is relatively inert and slowly absorbed, the osmotic gradient is maintained, thereby providing sustained ultrafiltration.

A number of randomized, controlled studies have shown that icodextrin is superior to hypertonic glucose solutions in improving ultrafiltration, especially among patients with high or high-average rates of peritoneal transport but also among patients with low-average peritoneal transport [37-44]. A systematic review of four randomized trials (102 patients) demonstrated improved ultrafiltration with icodextrin, with a mean difference of 449 mL/day (95% CI 289-608 mL/day) [45].

Although icodextrin is US Food and Drug Administration (FDA) approved only for once daily use, many clinicians use icodextrin for up to two exchanges daily [46-48]. Such use is considered off label.

Icodextrin has been associated with side effects including rash and culture-negative peritonitis (which were principally observed with certain product lots) (see "Peritoneal dialysis solutions", section on 'High-molecular-weight agents'). Patients with diabetes who use icodextrin need to be aware that icodextrin may cause falsely elevated blood sugar results with certain home-monitoring glucose testing strips, possibly leading to inappropriate therapy (see "Peritoneal dialysis solutions", section on 'High-molecular-weight agents'). Icodextrin metabolites may also interfere with serum assay for amylase activity. (See "Serum enzymes in patients with kidney failure", section on 'Amylase'.)

Additional exchange — We occasionally add an additional exchange during the long-dwell period if a trial of icodextrin is not available or was not helpful. The additional exchange is most likely to work in rapid transporters because of the shorter dwell time.

●For patients on CAPD, we break up the long overnight dwell into two equal parts using a cycler.

●For patients on APD who use daytime dwells, we use a midday exchange or a midday drain without further instillation of fluid, thereby resulting in a dry period.

Additional measures based on transport type — Additional measures are based on the transport type identified on the PET.

Fast (high) transporters — High transporters are treated with short dwells and occasionally resting the peritoneum.

●Dwell time – We shorten the dwell time of each exchange. Patients who develop hypervolemia on CAPD may benefit from being switched to APD, where the cycler performs three to five short-dwell exchanges during the sleeping period. However, the consequent long-day dwell may lead to fluid absorption if dextrose is used.

●Resting the peritoneum – In some patients, we temporarily switch to maintenance hemodialysis via a temporary central venous catheter in order to rest the peritoneum [49,50]. The strategy consists of the discontinuation of peritoneal dialysis for 4 to 12 weeks, with subsequent reinstitution of peritoneal dialysis. In a review of 33 patients treated by this technique, one single rest period allowed peritoneal function to return to earlier levels in 23 individuals (69 percent) [49]. The remaining patients required more than one rest period over several years.

In order to prevent adhesions in the peritoneum while on hemodialysis, we typically instill 100 to 200 mL of 1.36 percent glucose dialysate (1.5 percent dextrose) containing 3500 international units of heparin (which is allowed to dwell for 30 minutes and then drained) once or twice a week.

Slow (low) transporters — Patients with ultrafiltration failure and slow transporters (ie, type 3 ultrafiltration failure) are more difficult to treat. We try repeated dwells with 4.25 percent dextrose and use icodextrin for one long dwell, as described above. (See 'Icodextrin dialysate' above.)

These patients often need to change to hemodialysis or combined therapy (peritoneal dialysis and once or twice a week hemodialysis) to achieve euvolemia.

Increased lymphatic and postcapillary absorption — We shorten dwell times and use repeated dwells with 4.25 percent dextrose. We also use icodextrin for one dwell. (See 'Icodextrin dialysate' above.)

Occasionally, reducing the dwell volume is effective in reducing fluid absorption, particularly among smaller patients. Large-volume dwells may increase intraperitoneal pressure, which appears to increase absorption of fluid.

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

●Mechanism of fluid removal – Peritoneal dialysis removes fluid by ultrafiltration. Ultrafiltration is the passage of a fluid across a semipermeable membrane in response to a driving pressure (ie, hydrostatic, oncotic, or osmotic pressure). Ultrafiltration can be increased by increasing the amount of peritoneal dialysate dextrose or other solute such as icodextrin. (See 'Mechanism of fluid removal' above.)

●Prevention of volume overload – The most important measure to prevent hypervolemia is to establish and maintain an accurate target weight. It is also important to monitor residual kidney function and urine volume and the amount of fluid removed with dialysis. We use diuretics in all patients and attempt to preserve residual kidney function. (See 'Prevention' above.)

●Evaluation of volume overload – There are multiple causes of hypervolemia that may be related or unrelated to dialysis. We follow a stepwise and streamlined approach to determine cause. Our approach is defined above (algorithm 1). (See 'Epidemiology and causes' above and 'Evaluation' above.)

●Treatment of volume overload – Our approach to the treatment of hypervolemia is as follows:

•We restrict sodium to <2 grams/day. We restrict fluid to approximately 2 L/day if the patient does not have residual kidney function. (See 'Sodium and fluid restriction' above.)

•We treat all patients who have residual kidney function with increasing doses of diuretics. Furosemide may be given at doses up to 160 mg three times per day. Combination therapy with furosemide (200 mg orally twice daily) and a thiazide diuretic (eg, metolazone, 5 to 10 mg orally twice daily) may also be effective. (See 'Loop diuretics' above.)

•We use icodextrin-containing dialysate for one exchange per day. Icodextrin is better than hypertonic dextrose solutions in improving ultrafiltration. We may add an additional exchange during the long-dwell period if a trial of icodextrin is not helpful. (See 'Icodextrin dialysate' above and 'Additional exchange' above.)

•Additional measures are based on the transport type identified on the peritoneal equilibration test (PET). For fast transporters, we shorten the dwell time and often rest the peritoneum for 4 to 12 weeks. For slow transporters and for patients with increased absorption of fluid into lymphatics and local tissue, we try repeated dwells with 4.25 percent dextrose in addition to icodextrin for one long dwell. (See 'Additional measures based on transport type' above.)