Noninfectious complications of continuous peritoneal dialysis

INTRODUCTION — Some of the noninfectious complications that occur in patients on continuous peritoneal dialysis (eg, continuous ambulatory peritoneal dialysis [CAPD] and continuous cycler peritoneal dialysis [CCPD]) are due to increased intra-abdominal pressure resulting from instillation of dialysate into the peritoneal cavity. These include hernia formation, leaks (including hydrothorax or pleuroperitoneal leaks), local edema, back pain, and gastrointestinal problems, such as gastroesophageal reflux and delayed gastric emptying (see "Abdominal wall hernia and dialysate leak in peritoneal dialysis patients" and "Noninfectious complications of peritoneal dialysis catheters"). Complications of CAPD/CCPD not specifically related to intra-abdominal pressure include hemoperitoneum, pain on infusion or drainage of dialysate, electrolyte imbalances, and ultrafiltration failure.

Some of these problems will be reviewed here, including:

●Gastroesophageal reflux and delayed gastric emptying

●Back and abdominal pain

●Pleural effusion

●Hemoperitoneum

●Infusion and drain pain

●Hypokalemia

●Disorders of magnesium

Additional noninfectious complications are discussed separately:

●(See "Inadequate solute clearance in peritoneal dialysis".)

●(See "Abdominal wall hernia and dialysate leak in peritoneal dialysis patients".)

●(See "Noninfectious complications of peritoneal dialysis catheters".)

UPPER GASTROINTESTINAL SYMPTOMS — Nonspecific upper gastrointestinal symptoms are commonly reported by patients on continuous ambulatory peritoneal dialysis (CAPD) [1-3]. In a study that included 471 patients on CAPD without significant gastrointestinal disease, 387 (82 percent) had at least one gastrointestinal symptom [4]. Other studies of peritoneal dialysis report that nausea, fullness, and epigastric discomfort occur in as many as 20 to 50 percent of patients [1,2,5], while frequent vomiting occurs in approximately 14 percent [1,2].

When to suspect GERD or delayed gastric emptying — If the patient is well dialyzed (ie, has an adequate Kt/V) and other causes have been excluded, persistent upper gastrointestinal symptoms (eg, nausea, vomiting, fullness, or epigastric discomfort) should prompt consideration of gastroesophageal reflux disease (GERD) or gastroparesis. (See "Prescribing peritoneal dialysis" and 'Gastroesophageal reflux disease' below and 'Delayed gastric emptying' below.)

Among patients on peritoneal dialysis, there are many potential causes of these symptoms, including uremia, medication-induced gastroenteritis, gastritis, and peptic ulcer disease (see "Approach to the adult with nausea and vomiting" and "Clinical manifestations and diagnosis of gastroesophageal reflux in adults"). A paucity of data exists concerning the relationship between such symptoms and the presence of GERD or delayed gastric emptying (gastroparesis).

Gastroesophageal reflux disease — The general pathophysiology of GERD is discussed separately (see "Pathophysiology of gastroesophageal reflux disease"). Causes and treatment of GERD specific to peritoneal dialysis are discussed below.

●Increased intra-abdominal pressure and GERD – Increased intraperitoneal pressure is likely one of several rather than the sole cause of GERD in patients on peritoneal dialysis. Intraperitoneal pressures during peritoneal dialysis in the filled state average approximately 12 to 13 mmHg in the supine position and 21 to 22 mmHg in the erect position [6,7].

Intraperitoneal pressure rises linearly with increases in intraperitoneal volume [7]. However, there are conflicting data concerning the effects of increased intra-abdominal pressure on lower esophageal sphincter pressures and/or reflux [8,9]:

•In one study of 61 patients on peritoneal dialysis, a higher body mass index (BMI) was associated with higher intraperitoneal pressure and a higher incidence of enteric peritonitis, but there was no association with hernias, late peritoneal leaks, or GERD [7].

•In another study of 11 patients on CAPD, no systematic changes in intragastric or lower esophageal sphincter pressures, as determined by esophageal manometric studies, were observed in response to infusion of 1.5 to 2.5 liters of dialysate [8]. Although eight of these patients had chronic complaints of upper gastrointestinal symptoms, their findings on esophageal manometry were the same as those of the three who were asymptomatic.

•Another study of 13 patients on CAPD (four patients who had chronic nausea and vomiting and nine who were asymptomatic) also found no difference in supine lower esophageal sphincter pressures between the two groups after 2 liters of infused dialysate [9]. However, compared with asymptomatic individuals, symptomatic patients were observed to have a significantly higher number of reflux episodes (89 versus 17); reflux episodes lasting >5 minutes (2.3 versus 0.3); and an increased time in which the gastric pH was <4 (75 versus 11 minutes), as determined by 24-hour esophageal pH monitoring [9]. These data suggest that GERD is common in symptomatic patients on CAPD if the diagnosis is based upon 24-hour gastric pH monitoring rather than findings with esophageal manometry.

•A retrospective cohort study from Taiwan using a national database including nearly 1900 patients on peritoneal dialysis and 8955 patients on hemodialysis found that the risk for gastroesophageal reflux, intestinal obstruction or adhesions, and abdominal hernia was significantly higher among patients on peritoneal dialysis [10]. Even after propensity score, age, and sex matching, GERD was significantly more common among patients on peritoneal dialysis (hazard ratio 2.25, versus patients on hemodialysis) [10]. The authors implicate increased abdominal pressure with peritoneal dialysis as the mechanism for the observed higher risk of GERD in patients on peritoneal dialysis.

●Treatment – Medical management of GERD is similar to that for patients not on dialysis. (See "Medical management of gastroesophageal reflux disease in adults".)

In addition, the treatment of GERD in patients on CAPD/continuous cycler peritoneal dialysis (CCPD) may involve minimizing the intraperitoneal fluid volume. However, since achieving adequate dialysis in patients on CAPD/CCPD depends in large part on the exchange volume, this strategy may require careful tailoring of the dialysis prescription to attain adequate clearance while simultaneously minimizing symptoms of GERD. (See "Prescribing peritoneal dialysis".)

Delayed gastric emptying — Abnormalities of gastrointestinal motor function may lead to a delay in gastric emptying (gastric stasis), a disorder that occurs commonly in certain diseases, particularly diabetes [11]. (See "Pathogenesis of delayed gastric emptying".)

●Prevalence and mechanism – Delayed gastric emptying is relatively common among patients on CAPD, including those without diabetes. In a few small studies, delayed gastric emptying of solids and liquids has been detected in approximately 50 percent of both symptomatic and asymptomatic patients undergoing peritoneal dialysis [12-14].

Normal gastric emptying results have been noted when the peritoneal cavities are drained, thereby suggesting that a mechanical or neurogenic mechanism triggered by the presence of intra-abdominal fluid retards gastric emptying [12] (see "Gastroparesis: Etiology, clinical manifestations, and diagnosis" and "Pathogenesis of delayed gastric emptying"). However, this finding is not consistent across all studies [15,16]. In addition, certain types of dialysates (eg, glucose- or glycerol-based) may have an association with slower gastric emptying [16].

●Treatment – First-line therapy for gastroparesis in patients on CAPD/CCPD is oral metoclopramide. However, risks of central nervous system side effects, such as tardive dyskinesia, make this agent less attractive, particularly for long-term use, as is often required among patients with diabetes with gastroparesis. In general, the dose of oral metoclopramide should be reduced by 50 to 75 percent in patients with an estimated glomerular filtration rate (eGFR) <10 mL/min/1.73 m². An alternative to metoclopramide is oral erythromycin.

In patients whose symptoms persist in spite of oral therapy, intraperitoneal metoclopramide [17,18] or erythromycin [19,20] can be considered. However, the use of intraperitoneal metoclopramide has not been well studied in patients on CAPD/CCPD, and optimal dosing is unclear. One case report used 10 mg of metoclopramide per 2 liter bag of dialysate [17]. (See "Treatment of gastroparesis".)

Historically, the drug cisapride had been used to successfully treat delayed gastric emptying in patients on CAPD. However, it is no longer used in patients on dialysis due to the risk of arrhythmias associated with this drug.

In one study, two patients were successfully treated with gastric electrical stimulation, suggesting that this may also be considered in refractory cases [21].

BACK PAIN — Increased mechanical stress on the lumbar spine is common among patients on continuous ambulatory peritoneal dialysis (CAPD); this is due, in part, to the tendency of patients to assume a more lordotic position because of increased intra-abdominal pressure [22]. Some patients may also have poor abdominal muscle tone, resulting from previous surgeries or poor physical conditioning. These effects increase the mechanical stress on the lumbar spine, thereby causing back pain or sciatica [23]. Underlying degenerative disk disease, facet joint disease, and osteoporosis may further exacerbate the back pain [22]. (See "Evaluation of low back pain in adults".)

Treatment — Other than general therapeutic measures, the treatment of lower back pain in patients on CAPD may include decreasing the dialysate fill volume. However, since adequate dialysis must be maintained, changing the modality to continuous cycler peritoneal dialysis (CCPD) may be beneficial as larger fill volumes may be tolerated and the majority of dialysate dwells can be performed when the patient is supine. In those whose back pain is severe, transfer to hemodialysis may be necessary if adequate clearance on CCPD cannot be achieved. (See "Treatment of acute low back pain".)

PLEURAL EFFUSION DUE TO PLEUROPERITONEAL LEAK — The development of a pleural effusion (hydrothorax) in a patient on continuous peritoneal dialysis may be due to systemic volume overload, heart failure, or a local pleural process. The presence of a pleural effusion without other signs of heart failure or of peripheral edema, particularly if the effusion is only right sided, should prompt a search for a pleuroperitoneal communication.

Pathophysiology — A pleural effusion results from a pleuroperitoneal leak that is either congenital or acquired. A congenital defect typically occurs early in the course of therapy and is not related to the volume of instilled dialysate; as a result, congenital communications between the pleura and the peritoneum have been hypothesized to be the source of the leak [24-26]. Diaphragmatic hernias, for example, may allow dissection of fluid through defects around the major vessels and the esophagus or through diaphragmatic foramina [24,25]. In addition, the negative intrathoracic pressure, combined with an increased intra-abdominal pressure caused by dialysate instillation, may open small defects in the diaphragm and promote the flow of dialysate into the pleural space. In autopsy studies, discontinuities in the tendinous portions of the hemidiaphragms have been observed, thereby supporting the presence of diaphragmatic defects [25].

Acquired diaphragmatic defects also may also permit the development of hydrothorax. This is suggested by the finding that acute pleural effusion has been observed as late as eight years after starting peritoneal dialysis.

Epidemiology — The reported incidence of acute pleural effusion in patients being maintained on peritoneal dialysis varies from 1.6 to 10 percent [24,25,27,28]. In the largest study reported to date, 50 of 3195 patients (1.6 percent) developed acute hydrothorax after starting peritoneal dialysis [24]. Women are affected more commonly than men [27].

Patients with polycystic kidney disease are predisposed to develop pleuroperitoneal leaks. Because of the significantly reduced abdominal capacity of these patients, a marked increase in hydrostatic pressure occurs after the infusion of dialysate, thereby favoring the flow of dialysate from the peritoneum to the chest [25,29].

Clinical characteristics — Presenting symptoms and signs of pleural effusion include dyspnea and inadequate ultrafiltration ability [24]. However, many patients are asymptomatic (26 percent), with the recognition of fluid in the pleural space occurring only after routine physical examination [24]. The right side is more commonly affected than the left, with unilateral right-sided effusions found in 50 to 90 percent of cases [25,27]. Left-sided effusions may be less common because of the protective left diaphragmatic overlap of the heart and pericardium.

Most pleural effusions appear early after the initiation of peritoneal dialysis: 50 percent are found within 30 days of starting peritoneal dialysis, while 20 percent occur after one year [24]. Overall, these effusions have been reported to occur one day to eight years after beginning peritoneal dialysis [24,25,27].

Diagnosis — The presence of a pleural effusion resulting from a pleuroperitoneal leak should be suspected in the dyspneic nonedematous patient and in patients in whom ultrafiltration is consistently inadequate. Establishing that fluid in the pleural space is due to a pleuroperitoneal leak is based upon confirming that the fluid is dialysate and that it is leaking from the peritoneum into the pleura. A detailed discussion of the diagnosis and evaluation of this problem is presented separately. (See "Modalities for the diagnosis of abdominal and thoracic cavity defects in patients on peritoneal dialysis".)

Management — In most cases of peritoneal dialysis-related pleuroperitoneal leaks, the first step is cessation of peritoneal dialysis and drainage of any dialysate from the peritoneum. Acute thoracentesis to remove large pleural effusions is rarely required. Patients are then converted to hemodialysis for approximately four to eight weeks to allow the peritoneum to heal spontaneously. We do not attempt even low-volume peritoneal dialysis while resting the peritoneum since pleural effusions due to pleural leaks are unpredictable; they can rapidly increase in size and cause respiratory compromise. Unfortunately, the success rate with this conservative approach is low (approximately 40 percent) [24,25,27,30].

For patients who do not respond initially to the conservative approach above or who have recurrent pleural effusions, and who need or desire to continue with peritoneal dialysis, our approach depends on the availability of video-assisted thoracic surgery (VATS):

●For centers where VATS and experienced surgeons are available, we refer the patient for surgical repair of the pleuroperitoneal leak. VATS can be used both to identify the defective area and repair it using mesh, overlap suturing, glue, or sclerosing agents [25,31]. In this setting, VATS typically has a success rate >80 percent [32,33].

●If VATS and experienced surgeons are unavailable, chemical pleurodesis is a reasonable option. Pleurodesis involves placement of a thoracostomy tube under local anesthetic and the subsequent infusion of a sclerosing agent. Agents used for pleurodesis have included autologous blood, talc, and tetracycline [25,34,35]. Depending on the selected irritant, the procedure can be quite painful, and the effectiveness of such therapy is limited by the blind application of the chemical agent. Outcomes data for chemical pleurodesis in this setting are sparse; case series report success rates of approximately 50 percent [32,36]. (See "Management of nonmalignant pleural effusions in adults".)

Most patients with pleuroperitoneal leaks managed solely with conservative measures or who have chemical pleurodesis eventually switch to hemodialysis as their permanent mode of kidney replacement therapy [24,25,27,30].

HEMOPERITONEUM — A discussion of hemoperitoneum in patients undergoing peritoneal dialysis is presented separately. (See "Bloody peritoneal dialysate (hemoperitoneum)".)

PAIN WITH DIALYSATE INFUSION OR DRAINAGE — Abdominal pain often occurs with peritoneal dialysis-associated peritonitis (see "Clinical manifestations and diagnosis of peritonitis in peritoneal dialysis"). This symptom may also occur in the absence of peritonitis, frequently in association with infusion or drainage of dialysate.

Infusion pain — Many patients experience pain during infusion of dialysate into the peritoneal cavity that diminishes during the dwell period. This discomfort is frequently transient, resolving shortly after peritoneal dialysis is initiated [37,38]. In some patients, however, abdominal pain during dialysate infusion persists; in extreme situations, it may lead to discontinuation of peritoneal dialysis.

This complaint is thought to be caused by the acidic pH (pH 5.2 to 5.5) of conventional lactate dialysate. Additional contributing factors may include a poor catheter position (such as abutting against the bowel wall or peritoneal cavity surfaces), the dialysate temperature, and a high glucose concentration of hypertonic dialysis solutions [38].

●Treatment – Treatment of infusion pain includes the following measures, to be utilized in descending order [38]:

•Raising the dialysate pH, either by using specially formulated solutions or by injecting sodium bicarbonate (2 to 5 mEq/L) into standard dialysate prior to infusion

•Slowing the rate of infusion

•Injection of local anesthetics into the dialysis solution before infusing (1 percent lidocaine at 50 mg/exchange)

•Incompletely draining the fluid after a dwell period

•Catheter replacement

•Discontinuation of peritoneal dialysis

Although injecting sodium bicarbonate into the dialysis solution before infusion may be effective in alleviating pain [39], the risk of contamination during the injection with the subsequent development of peritonitis is high [38]. To avoid this risk, dialysis solutions with a higher pH (pH 7 to 7.4; such as bicarbonate and bicarbonate/lactate dialysis solutions) have been developed.

The effectiveness of such alternative solutions, compared with conventional lactate solution, has been evaluated in several studies [38,40]. In a randomized, double-blind, crossover study where pain was recorded using two quantifiable scales, significant reductions in inflow pain were noted when the bicarbonate (38 mM) or bicarbonate/lactate (25 mM of bicarbonate/15 mM of lactate) solutions were used as compared with the conventional lactate (40 mM solution) [38]. In this study, the bicarbonate/lactate solution was the most effective in alleviating infusion pain, thereby suggesting that factors other than the pH are involved in infusion pain.

Drain pain — Pain that occurs during the draining phase of peritoneal dialysis, or drain pain, is an underreported complication of peritoneal dialysis.

●Epidemiology – Drain pain is common. A Canadian multicenter study of 293 patients on automated peritoneal dialysis (APD) reported that 72 (25 percent) were using tidal peritoneal dialysis (TPD) due to drain pain [41]. A subsequent international study of 1630 patients on peritoneal dialysis reported that 461 (28 percent) had drain pain [42]. Among patients with pain, over a third rated their pain as severe.

Risk factors for drain pain include the use of APD (rather than continuous ambulatory peritoneal dialysis [CAPD]), polycystic kidney disease, and female sex [42].

●Etiology – Drain pain in patients on APD is thought to be caused by the cycler, which uses hydraulic pressure to drain dialysate. During drainage, this negative pressure may force the tip of the dialysis catheter against the bowel, bladder, or parietal peritoneum, causing pain. In some cases, it is possible that drain pain in APD may be the result of catheters that are too long and/or malpositioned. In such patients, the end portion of the catheter may lie on or near the floor of the peritoneal cavity, increasing the likelihood that the catheter will be suctioned into the parietal peritoneum during drainage, causing pain.

The higher rates of drain pain in patients with polycystic kidney disease and females may be due to diminished peritoneal space (from enlarged kidneys or the uterus) that increases the likelihood of the catheter being pushed into adjacent tissue.

●Clinical presentation – Drain pain predominantly affects patients on APD and presents as mild to severe discomfort in the abdomen or as referred pain in the rectum or perineum.

●Treatment – Our approach to the management of drain pain is based on our clinical experience and includes the following measures, to be utilized in descending order. If an intervention is effective, no further measures are required.

•Empiric treatment with laxatives. Successful treatment of even mild constipation may reposition the catheter and improve drain pain.

•Initiation of TPD. TPD leaves some peritoneal fluid in the abdomen with every cycle, which may reduce pain by preventing contact of the catheter with the viscera or parietal peritoneum. (See "Evaluating patients for chronic peritoneal dialysis and selection of modality".)

•Evaluation of the dialysis catheter with an abdominal radiograph. Patients who have a catheter that is too long and/or malpositioned should be referred to interventional nephrology/radiology or surgery. In some patients, catheter replacement or repositioning may improve or eliminate drain pain.

•Switching from APD to CAPD. CAPD may be effective reducing pain since it uses gravity rather than a cycler to drain dialysate.

•Discontinuation of peritoneal dialysis.

There is no evidence that adding bicarbonate or catheter design (straight versus coiled) has any impact on drain pain.

ELECTROLYTE ABNORMALITIES — Although any disturbance in electrolyte concentration may occur among patients on peritoneal dialysis, abnormalities in serum levels of potassium and magnesium are most commonly observed.

Hypokalemia — Unlike patients on chronic hemodialysis, in whom hyperkalemia is more common than hypokalemia, 10 to 35 percent of patients on continuous peritoneal dialysis require potassium supplements [43-45]. A retrospective study of 45 patients on continuous ambulatory peritoneal dialysis (CAPD)/continuous cycler peritoneal dialysis (CCPD) found that approximately 30 percent required oral potassium chloride supplementation with a mean dose of 22±13 mEq/day [43]. Patients with hypokalemia complained of weakness more often than those with normal potassium levels. African American race was a predictor of the need for potassium supplementation, but Kt/V, etiology of kidney failure, age, peritoneal membrane transport type, and oral protein and caloric intake were not. Cultural dietary preferences were thought to be responsible for these findings. A Chinese study of 886 patients on peritoneal dialysis found hypokalemia in 27.9 percent [46]. Cellular uptake of potassium, prompted by the intraperitoneal glucose load with subsequent insulin release, and bowel losses may also play a role in the hypokalemia observed in patients on peritoneal dialysis [47].

For stable chronic outpatients, liberalization of dietary potassium restriction and, when needed, oral potassium replacement (usually 20 mEq/day, based upon individual patient serum potassium determinations) are usually successful treatments for hypokalemia. Potassium-sparing diuretics like spironolactone may also be effective in the patient on peritoneal dialysis with chronic hypokalemia with appropriate surveillance of serum potassium levels [48]. Intraperitoneal potassium may also be administered acutely if critical hypokalemia exists [49]. Although intraperitoneal potassium can correct hypokalemia in stable outpatients on peritoneal dialysis [50], we recommend that oral potassium supplements be used rather than intraperitoneal potassium due to the risk of contamination (and subsequently peritonitis) when intraperitoneal instillation is required.

An association of hypokalemia with mortality has been observed among patients on peritoneal dialysis and suggests that hypokalemia should be recognized and treated appropriately [46,51]. The Chinese study of 886 incident patients on peritoneal dialysis found a U-shaped association of baseline serum potassium levels and all-cause and cardiovascular mortality, with the highest mortality among patients with potassium <3 mEq/L [46]. This observed relationship was only evident during the first year of dialysis and not in subsequent years.

A similar U-shaped relationship of serum potassium <3.5 mEq/L and all-cause and cardiovascular mortality was observed in a United States study that showed the lowest mortality in patients with serum potassium of 4 to 4.5 mEq/L [51]. A Brazilian study of 306 patients on peritoneal dialysis with hypokalemia (time-averaged serum potassium <3.5 mEq/L) compared with a control group of 1512 patients on peritoneal dialysis with normal time-averaged potassium levels also found a U-shaped relationship between time-averaged serum potassium and all-cause mortality [52]. In this study, hypokalemia was associated with a 49 percent increased risk for cardiovascular mortality [52]. Hypokalemia was also associated with a higher mortality due to nonperitoneal dialysis-related infections; there was no association of hypokalemia with peritonitis [52]. Increased mortality in patients on peritoneal dialysis with hypokalemia was demonstrated to be independent of the use of potassium supplementation in a study of 108 patients receiving potassium supplements compared with 114 control patients [53].

Hypermagnesemia and hypomagnesemia — Hypermagnesemia, a common finding in patients on peritoneal dialysis, is due to positive magnesium balance resulting from kidney failure and the relatively high dialysate magnesium concentration (see "Regulation of magnesium balance" and "Hypermagnesemia: Causes, symptoms, and treatment"). Serum magnesium levels are usually elevated in those dialyzed against solutions containing magnesium concentrations of 0.75 mmol/L (1.8 mg/dL) [54,55].

The typical serum magnesium level in patients with end-stage kidney disease (ESKD) is 2.4 to 3.6 mg/dL (1 to 1.5 mmol/L), a value usually not associated with clinical symptoms (see "Hypermagnesemia: Causes, symptoms, and treatment"). However, an inverse relationship between concentrations of magnesium and intact parathyroid hormone (PTH) raises the possibility that hypermagnesemia suppresses PTH levels, thereby contributing to adynamic bone disease in patients on peritoneal dialysis [54]. As a result, although abnormalities in calcium, vitamin D, and phosphorus metabolism are the primary factors, more attention should be focused upon the role of magnesium in leading to the relative high incidence of adynamic bone disease in patients on peritoneal dialysis. (See "Overview of chronic kidney disease-mineral and bone disorder (CKD-MBD)".)

In addition, multiple factors can affect plasma magnesium concentration including diet, nutritional status, albumin levels, medications like proton pump inhibitors, magnesium-containing laxatives, and dialysis prescriptions [56]. Little is known about hypo- and hypermagnesemia among patients on dialysis, especially on cardiovascular status and mortality, and more study of this cation in relation to dialysis is needed [56,57].

Current peritoneal dialysis solutions contain magnesium concentrations of 0.75, 0.5, and 0.25 mmol/L (1.8, 1.2, and 0.6 mg/dL). As some patients utilizing the 0.25 mmol/L (0.6 mg/dL) magnesium concentration become hypomagnesemic, the 0.5 mmol/L (1.2 mg/dL) concentration dialysate may be the better choice [55].

Magnesium clearance in patients on peritoneal dialysis likely also varies, in part, based on the presence or absence of residual kidney function and the characteristics of the individual's peritoneal membrane [58]. In one study of 115 patients on peritoneal dialysis, for example, peritoneal magnesium clearance positively correlated with nutritional status, daily peritoneal protein loss, and high transport membrane type [58].

Given that many patients on peritoneal dialysis are slightly hypermagnesemic, the use of magnesium-containing antacids as phosphate binders is rarely a useful clinical option.

Studies have demonstrated an association of mortality with hypomagnesemia among patients on peritoneal dialysis [59,60]. A single-center, retrospective study of 253 incident patients on peritoneal dialysis found that 14.2 percent were hypomagnesemic [59]. Peritoneal dialysis duration and hypoalbuminemia were associated with low magnesium levels in these patients who also had a higher mortality and cardiovascular mortality [59]. A large cohort study of 10,692 incident patients on peritoneal dialysis beginning peritoneal dialysis between January 2007 and December 2011 found that 18 percent of patients had a serum magnesium level <1.8 mg/dL, 21 percent had a magnesium level of 1.8 to 2.0 mg/dL, and only 19 percent had a level ≥2.4 mg/dL [60]. Patients with magnesium level <1.8 mg/dL had a higher risk of hospitalization (HR 1.23) and a higher risk of death (HR 1.21). The greatest risk for hospitalization was in hypomagnesemic patients with a serum albumin <3.5 mg/dL [60]. Thus, emerging evidence suggests that up to a quarter of patients on peritoneal dialysis are hypomagnesemic and that hypomagnesemia may be a mortality risk factor in this patient population. Additional study of these issues is needed.

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".)

INFORMATION FOR PATIENTS — UpToDate offers two types of patient education materials, "The Basics" and "Beyond the Basics." The Basics patient education pieces are written in plain language, at the 5^th to 6^th grade reading level, and they answer the four or five key questions a patient might have about a given condition. These articles are best for patients who want a general overview and who prefer short, easy-to-read materials. Beyond the Basics patient education pieces are longer, more sophisticated, and more detailed. These articles are written at the 10^th to 12^th grade reading level and are best for patients who want in-depth information and are comfortable with some medical jargon.

Here are the patient education articles that are relevant to this topic. We encourage you to print or e-mail these topics to your patients. (You can also locate patient education articles on a variety of subjects by searching on "patient info" and the keyword(s) of interest.)

●Beyond the Basics topics (see "Patient education: Peritoneal dialysis (Beyond the Basics)")

SUMMARY AND RECOMMENDATIONS

●Noninfectious complications of peritoneal dialysis – Noninfectious complications occurring in patients undergoing either continuous ambulatory peritoneal dialysis (CAPD) or continuous cycler peritoneal dialysis (CCPD) are often a consequence of increased intra-abdominal pressure resulting from instillation of dialysate into the peritoneal cavity. These include gastroesophageal reflux and delayed gastric emptying, back and abdominal pain, and pleural effusion. Additional common issues observed in patients on peritoneal dialysis include hemoperitoneum, pain with dialysate infusion or drainage, hypokalemia, and dysmagnesemia. (See 'Introduction' above.)

●Gastrointestinal complications – Gastroesophageal reflux and delayed gastric emptying should be considered in well-dialyzed patients complaining of nausea, vomiting, fullness, and epigastric discomfort. In addition to general measures, the treatment of gastroesophageal reflux in patients on CAPD/CCPD may involve minimizing the intraperitoneal fluid volume. However, since achieving adequate dialysis in patients on CAPD/CCPD depends in large part on the exchange volume, this strategy may require careful tailoring of the dialysis prescription to avoid inadequate small-solute clearance, especially since there is no clear association of increased intra-abdominal pressure with the occurrence of gastroesophageal reflux disease (GERD) in small studies of patients on peritoneal dialysis. (See 'Upper gastrointestinal symptoms' above.)

●Pleuroperitoneal leak – A pleuroperitoneal leak should be considered in patients on peritoneal dialysis with pleural effusion (especially right-sided effusion) without other signs of heart failure. A number of different measures can be utilized for treatment. (See 'Pleural effusion due to pleuroperitoneal leak' above.)

●Pain with dialysate infusion or drainage – Abdominal pain in a patient on peritoneal dialysis should prompt consideration of peritonitis, catheter malfunction, or possibly infusion or drain pain. Infusion pain may be treated with intraperitoneal infusion of bicarbonate. We take a stepwise approach to the management of drain pain. (See 'Pain with dialysate infusion or drainage' above.)

●Hypokalemia – Unlike patients on hemodialysis, hypokalemia is fairly common in patients on peritoneal dialysis and is likely associated with all-cause and cardiovascular mortality in a U-shaped manner. Liberalization of dietary potassium, and often prescription of oral potassium supplements, may be needed. Medications that promote hyperkalemia such as the potassium-sparing diuretics (eg, spironolactone) may be useful. (See 'Electrolyte abnormalities' above.)

●Magnesium disorders – Although hypermagnesemia is the predominant problem in patients with kidney failure, hypomagnesemia has been observed in a significant percentage of patients on peritoneal dialysis and may be associated with hospitalization and mortality. Magnesium levels should be monitored and dialysate magnesium concentration adjusted to maintain normal levels. (See 'Hypermagnesemia and hypomagnesemia' above.)

ACKNOWLEDGMENT — The UpToDate editorial staff acknowledges Jean L Holley, MD, FACP, who contributed to earlier versions of this topic review.