In dialysis patients, solute clearance is determined by clearance derived from both dialysis therapy and RRF. When we examined solute clearance of dialysis patients, we have to consider total solute clearance, the sum of clearance delivered by dialysis and RRF . In the early period of dialysis, most patients are likely to have sufficient RRF , which lead to significant contribution to total solute clearance. As dialysis duration is prolonged, RRF tends to decline. Therefore, patients may require more solute clearance via dialysis, and prescribing greater dose of dialysis may be necessary over time. Previous clinical studies and guidelines indicated a minimum or “adequate” solute clearance target. If this solute clearance target is achieved, further adjustment of dialysis dose may depend on various essential indicators for optimal dialysis.

Volume overload is significantly implicated with adverse cardiovascular disease including congestive heart failure , left ventricular hypertrophy, and hypertension; therefore, it is important to monitor ultrafiltration volume , dry weight , sodium intake, and other clinical assessments of volume status (Peritoneal Dialysis Adequacy Work Group 2006). In addition, several studies have indicated that volume overload is closely implicated with mortality in PD patients, irrespective of symptom and signs of volume overload (Blake 1997). Although there was no published evidence, the EBPG committee suggested an arbitrary target ultrafiltration goal of 1.0 L/day to increase the awareness of the maintenance of a euvolemic state (Dombros et al. 2005a). Afterward, the work group for the clinical practice guidelines on PD adequacy of Canadian Society of Nephrology suggested the more detailed guidelines on the management of hypervolemia. They suggested that a low net daily peritoneal ultrafiltration volume (<750 mL in anuric patients or <250 mL in patients with RRF ) would be an indication for careful evaluation of volume status (looking for evidence of volume overload) and of dietary fluid and food intake (looking for evidence of insufficient intake or malnutrition) (Blake et al. 2011). In contrast, some experts did not suggest a minimum target of ultrafiltration volume, because aggressive ultrafiltration may facilitate RRF decline. In accordance with the 2006 NKF-K/DOQI guidelines (Peritoneal Dialysis Adequacy Work Group 2006), we propose that optimal amount of ultrafiltration should be tailored by individual volume status and blood pressure levels rather by arbitrary goal. In addition, we need to monitor the nutritional status such as the amount of food and water intake in patients without any sign of volume overload in spite of an ultrafiltration volume of <1.0 L/day.

Nutrition is an important predictor of clinical outcomes in dialysis patients whether on HD or on PD. A large number of factors have been identified as the causes of malnutrition in dialysis patients, which include anorexia, uremia derived from insufficient dialysis, advanced age, chronic inflammation , metabolic acidosis , impaired anabolism, gastrointestinal or medical comorbidities, and socioeconomic problems (Blake et al. 2011). Notably, in PD, protein loss to dialysate may happen, resulting in increased risk of malnutrition. Estimated amount of protein and amino acid loss is reported to be up to 15 g and 2–4 g per day, respectively (Bergstrom et al. 1993). Therefore, assessment of nutritional status should be performed as a routine practice in PD patients. The EBPG recommended to use subjective global assessment (SGA), protein intake derived from the protein equivalent of nitrogen appearance (PNA) or dietary recall, and an assessment of protein nutrition in PD patients every 6 months (Dombros et al. 2005b). Multidisciplinary team including physician, nurse, dietician, social worker, and family members is necessary for comprehensive assessment and effective management of malnutrition . If malnutrition of the patients is likely to be associated with underdialysis, the adherence of the patients to the PD prescription and the delivered dose of PD should be reevaluated at the time and adjusted to a sufficient level.

Routine evaluation for uremia in PD patients is required for determining optimal dialysis. Patients may be asked if they have intractable anorexia , vomiting , or itching for the assessment of uremic symptoms . In addition to uremic symptoms , pericardial effusion, confusion suggestive of uremic encephalopathy, and gum or mucosal bleeding are also indications to deliver more dialysis dose even the minimum target Kt/Vurea of 1.7 is achieved.

Prevention of hyperkalemia is crucial in dialysis patients. Because PD solution does not contain potassium and potassium is constantly removed by diffusion, hyperkalemia is less likely to occur in PD patients who maintain adequate dialysis. Therefore, PD patients with severe hyperkalemia should be evaluated whether they had drugs causing hyperkalemia including renin-angiotensin receptor blockers, nonsteroidal anti-inflammatory drugs, nutritional supplements, and herbal remedies or whether they skip PD exchanges. Metabolic acidosis has been regarded to induce harmful complications such as malnutrition , inflammation , and mineral bone disease (Kraut and Kurtz 2005). However, only 25% of CAPD patients achieve normal serum bicarbonate value by using a lactate solution (Feriani 1996). Although there is no consensus regarding the optimal serum bicarbonate levels in PD patients, individualizing PD solution using different buffers may be helpful to correct metabolic acidosis (Feriani et al. 1997; Feriani et al. 2004). To date, there is growing attention to mineral metabolism and patient outcome in PD patients (Noordzij et al. 2006). Standard solution containing 3.5 mEq/L calcium may be associated with hypercalcemia especially in patients with high-dose calcium-based phosphate binders. Even though optimal calcium concentrations of PD solution remain unclear, adjusting dialysis solution and phosphate binders is necessary in patients with hypo-/hypercalcemia .

RRF is independently associated with clinical outcomes in dialysis patients. It is recognized that PD preserves RRF better than HD at the initiation of dialysis . RRF promotes urea or creatinine clearance , helping to achieve target Kt/Vurea . If a patient has a creatinine clearance of 1 mL/min and weekly creatinine clearance reaches 10 L/week, it is a very considerable value which we should do our best to preserve. RRF also increases fluid excretion, leading to mitigation of adverse effect derived from chronic volume overload (Bragg-Gresham et al. 2007). Furthermore, greater clearance of middle molecule such as β2-microglobulin and indoxyl sulfate (Bammens et al. 2003), better phosphate and anemia control (Penne et al. 2011), and increased dietary protein and calorie intake representing better nutritional status (Wang et al. 2001) were significantly associated with RRF. To preserve RRF, all PD patients are indicated to take renin-angiotensin receptor blockers, even they are not hypertensive. Two randomized controlled study demonstrated that angiotensin receptor blockers (ARB) or angiotensin-converting enzyme inhibitors (ACE inhibitor) preserved RRF in PD patients (Li et al. 2003; Suzuki et al. 2004). In addition, although blood pressure requires strict control, excessive ultrafiltration should be avoided due to harmful effect of volume depletion (Table 14.2).

Since PD is a home-based therapy, it usually requires a daily performance of dialysis and clinical assessment of patient’s condition by a patient and/or a caregiver. Therefore, adherence to therapy is a critical issue in PD. However, according to a systematic review investigating nonadherence in PD patients, rates of missing PD exchanges or sessions were reported to be 2.5–53%, and shortening of cycles was reported in 4–15% of patients (Griva et al. 2014). Moreover, over half of the studies (13 of 20 studies, 65%) indicated the rates of nonadherence to PD regimen to be higher than 20%. Previous studies have demonstrated that non-compliance to dialysis therapy increased the risk of mortality and hospitalization in PD patients (Bernardini et al. 2000; Bernardini and Piraino 1998). In this regard, clinician should pay attention to patient’s adherence and adjust dialysis regimen.

Each of the patients starts PD for a wide variety of reasons. PD prescription should be tailored in consideration of these reasons and circumstances. However, in real world, it is impossible to fulfill all of the aforementioned indicators for optimal dialysis at the first prescription of PD. For practical application, we categorized patients into three groups according to the chronicity of kidney disease and planned start of dialysis, (1) acute PD in patients with acute kidney injury (AKI), (2) urgent start of PD in newly diagnosed chronic kidney disease (CKD) patients, and (3) planned start of PD in known CKD patients, and suggest empirical prescription of the first PD. Nevertheless, we should keep in mind that these empirical prescriptions may not be appropriate for all patients and prescription can be changed for individual goal of optimal dialysis.