Clinical practice guidelines for the provision of renal service in Hong Kong: Peritoneal Dialysis

Guideline 1.1

There is a considerable variation in the scope of service provided by different PD units, and it is unrealistic to dictate a uniform space requirement. In general, PD unit should encompass dedicated PD training rooms, store rooms, clean and dirty utility rooms, clinic area, access to emergency beds and HD, toilet and showers, as well as office for nurse, doctors, clerical and administrative staff.1

Guideline 1.2

Good patient training is crucial for the success of all PD programmes.2 A dedicated area, preferably in the form of a separate room, would facilitate the training procedure and enhance infection control. PD training area should generally include the following equipment: comfortable chairs and beds, wash basins, surface or trolley, weighing scales, drip stands or hooks, shelving for consumables, bag-warming equipment, ambulatory PD machine, clock and sphygmomanometer.1

Guidelines 2.1 to 2.3

The Department of Health of Hong Kong government has a detailed set of regulation is this regard and should be followed.3 In general, standards recommended by the Renal Association Standards Subcommittee of the Royal College of Physicians should also be followed.4

Guideline 3.1

Inadequate staffing ratios of physicians or nurses to patients are likely to be associated with worse clinical outcome.5 In addition, achieving good clinical governance requires a clearly defined hierarchy of responsibility and an appropriate mix of technical, nursing and medical personnel.6 The presence of a dedicated nephrologist in a PD programme is of great importance to guarantee success.7

Guideline 3.2

Clinical governance is an essential component of a successful PD programme. Regular unit meetings facilitate accountability and internal communication between staff of various expertise.8 In addition, continuous quality improvement (CQI) programmes are effective in quality assurance as well as improving the outcome of PD patients.7, 9

Guideline 3.3

Continuous quality improvement programme within a PD unit may help to reduce peritonitis rate.10 Medical staffs who are committed to the PD programme need to assume leadership roles to ensure professional competence and confidence in PD by establishing appropriate protocols and training curricula, and mandating the provision of continuous education.6 The latest International Society for Peritoneal Dialysis (ISPD) guideline recommends that multidisciplinary teams running CQI programmes in PD centres should meet and review their units’ performance metrics regularly.11

Guideline 4.1

Creating procedures and training protocols is a critical element of a successful PD programme.12, 13 Infrastructure deficiencies have been proposed as the major reason of unsuccessful PD programme in USA.14

Guideline 4.2

The smooth running of a PD programme should be the shared responsibility of the nephrologist in-charge and the PD nurse manager.1 Preparation of procedure protocols is one of the first steps in PD programme development.13, 15, 16 In general, protocols should be evidence-based whenever possible. After the PD programme is successfully underway, periodic review and revision to the protocols are necessary to reflect evolution of clinical practice and to meet regulatory requirements.17

Guideline 5.1

Multiple studies show that late referral to nephrologist is associated with adverse clinical outcome of dialysis patients.18 A timely referral means referral to nephrology service at least 1 year before start of dialysis,19 which allows adequate pre-dialysis education. Pre-dialysis education can be accomplished by a variety of methods, but should be based on principles of adult education with a chronic disease focus.20 Individual counselling is appropriate for all patients. However, group classes may be helpful for the patient and family members, and allow utilization of staff time more efficiently. Pre-dialysis education programmes are often rewarding and may be valuable for all prospective dialysis patients.20

Guidelines 6.1 and 6.2

Uraemic symptoms often but not invariably occur in the eGFR range between 5 and 10 mL/min per 1.73 m^2.19 Survival advantage of early start of dialysis was not confirmed by published trials.21, 22 For example, in the Initiating Dialysis Early and Late Study (IDEAL) study, asymptomatic patients started on dialysis at eGFR 5–7 mL/min per 1.73 m² had similar survival with those dialyzed at eGFR 10–14 mL/min per 1.73 m^2.21 Given the risks and benefits of dialysis, as well as the potential imprecision of measurements, patients need to be treated according to symptoms and signs, not simply based on a laboratory value.19 On the other hand, some patients are more susceptible to uraemic symptoms, and they may require dialysis at a higher eGFR.

Guideline 7.1

There is no ideal method or personnel for PD catheter insertion. Open mini-laparotomy, blind trocar technique, peritoneoscopic or laparoscopic implantation has all been reported to be successful.23, 24 With appropriate treatment, surgeons, urologists and nephrologists are all capable of performing the procedure.25, 26 Irrespective to the method and personnel, however, the ISPD clinical practice guidelines for peritoneal access should be followed.27 Nephrologists who perform PD catheter insertion should have the appropriate credentialing as recognized by the Hong Kong College of Physicians.

Guideline 7.2

At least four randomized control trials show that a single dose of preoperative antibiotic given intravenously before PD catheter insertion reduces early peritonitis, but not exit-site and tunnel infection.28, 29 A first-generation cephalosporin (e.g. cefazolin) is the most frequently used agent for this purpose.30 Vancomycin is the appropriate alternative in case of penicillin allergy or documented carrier of methicillin-resistant Staphylococcus aureus (MRSA).

Guideline 8.1

Unless there is a pressing clinical need, the wound dressing after PD catheter insertion should be kept undisturbed for at least 48 h in order to allow re-epithelialization. During the process of abdominal wound healing, fibrous tissue deposition takes place from 72 h to 4 weeks.31 The European best practice guidelines32 and ISPD guidelines27 both state that the time between catheter insertion and CAPD beginning should be at least 2 weeks to avoid early leakage.

Guideline 8.2

Recent studies showed that immediate PD after catheter insertion is feasible.33, 34 However, PD with a low volume (1 L cycles for adult patient) in the supine position should be used to minimize mechanical stress to the wound and the risk of dialysate leak.27, 32

Guideline 9.1

High-quality evidence that guides how, where, when and by whom PD training should be performed, however, is lacking. While there are no studies evaluating the education or abilities of the trainer, nursing staff usually take a leading role in coordinating the efforts of the PD team to provide home care for the patients,36 and are therefore more likely to possess the qualities required for patient training.35 To successfully teach patients, the nurse must acquire knowledge on education theories and related practical skills.1 In addition to the initial training, home visit by PD nurses is often useful in detecting practical problems.37 Re-training in selected patients may also be beneficial.11 In general, the latest ISPD recommendations for teaching PD patients and their caregivers should be followed.38

Guideline 9.2

A formalized programme is the best method to prepare a patient for self-management of any chronic disease.39 As mentioned in guideline #4.2, a structured protocol, or, in the case of patient training, a detailed curriculum is essential for ensuring the quality of training.1, 13, 15, 16 Patient education should be documented.40

Guideline 10.1

In order to make sensible PD prescription and optimize solute clearance and fluid removal, it is important to determine the peritoneal membrane transport characteristics of each PD patient.41, 42 Baseline peritoneal transport characteristics are particularly important because once established, these data can be used to guide prescription writing and predict clearances and ultrafiltration volumes.43, 44 Kinetic modelling programmes have been developed that use data from the standard PET to help in prescription management.45, 46 However, PET should not be performed sooner than 2 weeks after dialysis commencement because of unstable peritoneal permeability at this stage.42 The standard PET with 2.5% dextrose solution as described by Twardowski is generally preferred.47 The modified PET with 4.25% dextrose solution is an acceptable option when ultrafiltration failure is suspected.48

Guideline 10.2

Peritoneal transport characteristics may change over time.49-52 Prolonged PD therapy in itself leads to progressive change in peritoneal transport and ultrafiltration capacity,49, 50 which is often exacerbated following an episode of severe peritonitis.51 After an episode of severe peritonitis that requires catheter removal and temporary HD, PET should be repeated because peritoneal transport may change drastically and ultrafiltration failure is common.52 PET should also be repeated when there is clinical suspicion of ultrafiltration failure. In these situations, PET may help to tailor further PD regimen.

Guideline 10.3

PET is not reliable during and shortly after peritonitis episodes. Peritonitis causes peritoneal hyperaemia, which results in transient increase in peritoneal transport of low molecular weight solutes, increase in rates of glucose absorption, and reduction in ultrafiltration.53 These changes usually resolve within a month after resolution of the peritonitis.54, 55

Guideline 11.1

Both local and international studies showed that total Kt/V(urea) below 1.7 is associated with poor clinical outcome of patients treated with CAPD.56, 57 On the other hand, achieving total Kt/V(urea) greater than 1.7 does not result in additional clinical benefits.58, 59 Although renal and peritoneal clearances are probably different biologically, there is no published evidence to guide the minimal target of peritoneal Kt/V. Published data in patients receiving machine-assisted PD are more limited. It is generally believed that the target Kt/V(urea) for machine-assisted PD should be somewhat higher than CAPD, but the exact figure remains to be defined.

Guideline 11.2

There is no available evidence as to the optimal frequency of dialysis adequacy monitoring for patients on PD. The current recommendation strikes a balance between international guidelines41, 42 and practical feasibility in Hong Kong. In addition to the minimal recommendation of yearly measurement, dialysis adequacy measurement should be considered when there is clinical suspicion of under-dialysis, sudden change in peritoneal transport characteristics (e.g. after an episode of severe peritonitis that require catheter removal), or rapid loss of residual renal function.

Guideline 12.2

Since residual renal function is a strong prognostic indicator for PD patients, its preservation should be a major objective in the management of PD patients.41 Two randomized control trials consistently showed that the use of ACE inhibitors or ARBs is associated with a reduction in rate of residual renal function decline.63, 64 The use of these agents is therefore recommended when antihypertensive therapy is indicated for PD patients.41 The use of biocompatible PD solution may also help to preserve residual renal function,65 but the magnitude of effect is small. There is no definitive evidence that exposure to aminoglycoside accelerates the loss of residual renal function.66 Nonetheless, nephrotoxic drugs should be avoided.

Guideline 13.1

Blood pressure and bodyweight measurements are an integral part of primary medical care. Most PD patients are hypertensive; accurate assessment of blood pressure is crucial for patient care.67 Although office blood pressure may not reliably reflect the overall blood pressure control of a patient,68 it is the parameter used in almost all antihypertensive trials. Regular self-measurement of blood pressure at home is strongly encouraged. Short-term fluctuations in bodyweight (in days to weeks) often represent change in body fluid status, while the long-term trend (in months) likely reflects alteration in body built and nutritional status.69 Since blood pressure and bodyweight are non-invasive, they should be measured and documented in every clinic visit.

Guideline 13.2

There is no published data as to the optimal frequency of serum biochemistry monitoring in PD patients. The Caring for Australasians with Renal Impairment (CARI) Guidelines recommend monitoring at least every 2 months.42 Since most of the stable PD patients in Hong Kong are followed every 2–3 months, our current recommendation represents a compromise between optimal ideal and our local practice. The exact panel of biochemical test depends on local resource as well as individual patient need, but usually includes serum or plasma sodium, potassium, urea, creatinine, albumin, liver enzymes, calcium and phosphate. Fasting plasma glucose and glycated haemoglobin (HbA1c) levels should be measured at least half yearly for diabetic patients. Fasting plasma or serum cholesterol, triglyceride, PTH level and iron profile are also commonly monitored at less frequent intervals.

Guideline 13.3

There is no published data as to the optimal frequency of PTH level monitoring in PD patients. The CARI Guidelines recommend monthly checking when there are changes of therapy that may influence PTH, and every 2–3 monthly in other patients,70 which may not be possible in Hong Kong. We acknowledge the deficiencies in our current recommendation, which is largely governed by the financial resource and practical feasibility in Hong Kong.

Guideline 14.1

Anaemia following dialysis initiation has been shown to be associated with increased mortality in PD patients.71 International guidelines recommend a very tight Hb range of between 11.0 and 12.0 g/dL.72, 73 However, maintaining Hb levels in such a narrow range is difficult in practice. The CARI guidelines currently recommends a target Hb level of 10.0–11.5 g/dL,74 which seems a reasonable range, with the consideration of published literature as well as clinical reality.

Guideline 14.2

For any patient who may require ESA therapy, the potential benefits must be balanced with the clinical risks. Iron deficiency (e.g. due to occult blood loss), haemolysis, and other secondary causes of anaemia should be excluded. In anaemic PD patients, the objectives of ESA treatment are to avoid blood transfusion, especially in patients awaiting kidney transplantation, and to improve the quality of life (QOL).74 There is substantial inter-individual variation in the Hb level that would lead to anaemic symptoms, but non-specific symptoms in patients with Hb above 9.5 g/dL are unlikely caused by anaemia and ESA would not be beneficial. Patients with underlying coronary artery disease may require a slightly higher haemoglobin level to avoid angina. On the other hand, ESA treatment may be withheld for a lower haemoglobin level in patients with a history of malignancy for the possibility of its trophic effect on tumour growth.

Guideline 14.3

There is no published data as to the optimal frequency of Hb monitoring in PD patients. The CARI Guidelines recommend monitoring at least every 1–3 months.74 Since Hb level is usually an integral part of regular blood test, we recommend it should be monitored simultaneously with serum biochemistry and at the same frequency.

Guideline 15.1

A previous observational study showed that CQI programmes with regular monitoring effectively reduces peritonitis rate.9 This practice is endorsed by the latest ISPD recommendations.11 Peritonitis rate should be reported as the number of episodes per patient-year rather than number of patient-month per episode,11 which was commonly used in the past. In addition to the overall peritonitis rate, PD centre should ideally monitor the peritonitis rates of specific organisms, the percentage of patients per year who are peritonitis free, and the antimicrobial susceptibilities of the infecting organisms.11 Organism-specific peritonitis rates should also be reported as absolute rates, that is, as number of episodes per year. Sampling and culture methods of PD effluent should be reviewed and improved if more than 15% of peritonitis episodes are culture-negative.11

Guideline 15.2

Patients with peritonitis usually present with cloudy fluid and variable degree of abdominal pain.11, 76 For the purpose of unit audit, peritonitis should be diagnosed when at least 2 of the following are present: (i) clinical features consistent with peritonitis, that is, abdominal pain and/or cloudy dialysis effluent; (ii) dialysis effluent white cell count greater than 100/μL or greater than 0.1 × 10⁹/L (after a dwell time of at least 2 h), with greater than 50% polymorphonuclear neutrophil and (iii) positive dialysis effluent culture.11 However, cloudy PD effluent almost always represents infectious peritonitis although there are other causes.11, 77, 78 Since early effective treatment is critical for therapeutic success, PD patients presenting with cloudy effluent should be presumed to have peritonitis and treated as such until the diagnosis can be confirmed or excluded. When a patient presents with cloudy PD effluent or whenever peritonitis is suspected, PD effluent should be tested for cell count, differential, Gram stain and culture. Blood-culture bottle is the preferred technique for bacterial culture of PD effluent.11 Centrifuging PD fluid and culturing the pellet, or the lysis centrifugation technique may further improve the diagnostic yield. There is insufficient evidence to currently support the use of novel techniques for the diagnosis of peritonitis.

Guideline 15.3

There are potentially serious consequences of peritonitis (relapse, catheter removal, permanent transfer to HD and death), which are more likely to occur if treatment is not promptly initiated. To prevent delay in treatment, antibiotic therapy should be initiated as soon as cloudy effluent is seen, without waiting for confirmation of the cell count from the laboratory.11

Guideline 15.4

Since delayed effective treatment is associated with serious consequences (see above), it is important for the antibiotic protocol to cover all pathogens that are likely to be present. No antibiotic regimen has been proved to be superior to the others as empirical treatment. The selection of empiric antibiotics must be made in light of both the patient's and the programme's history of microorganisms and sensitivities.79 In general, Gram-positive organisms may be covered by vancomycin or a cephalosporin, and Gram-negative organisms by a third-generation cephalosporin or aminoglycoside.11 Vancomycin is often not preferred for empirical Gram-positive coverage because of the worry of inducing vancomycin resistant organisms, except in PD units with a high prevalence of methicillin-resistant bacteria. Aminoglycosides and ceftazidime have similar efficacy as empirical Gram-negative coverage. There is no evidence that short courses of aminoglycosides accelerate the loss of residual renal function, but repeated or prolonged aminoglycoside treatment (more than 3 weeks) should be avoided.

Intra-peritoneal antibiotics are the preferred route of administration unless the patient has features of systemic sepsis.11 IP aminoglycoside should preferably be administered as daily intermittent dosing. IP vancomycin should be administered intermittently every 4–5 days. Since the dosage interval varies with bodyweight and residual renal function, measurement of serum vancomycin level may help the clinical decision, and serum vancomycin level should be kept above 15 μg/mL.11 IP cephalosporin may be administered either continuously or on a daily intermittent basis, although continuous dosage is preferred for pharmacokinetic considerations.

Guideline 15.5

Within 48 h of initiating therapy, most patients with PD-related peritonitis should have considerable clinical improvement.11 It is good clinical practice to avoid continuing unnecessary broad spectrum antibiotics in order to avoid the emergence of resistant organisms.80

Guideline 15.6

Although there is no randomized control trial in this area, clinical experience and the latest ISPD guideline recommend treatment of 2 weeks as the minimal duration.11 Published series suggest that peritonitis episodes caused by Staphylococcus aureus or Pseudomonas species have an increased risk of developing relapsing episodes and may be benefited from a longer duration of therapy.81, 82 Patients who responded slowly to antibiotics, as well as those with severe peritonitis, especially episodes caused by Enterococcus species or mixed bacterial growth, may also be benefited from a longer course of treatment.

Guideline 15.7

The focus of peritonitis treatment should always be on preservation of the peritoneum rather than saving the peritoneal catheter.11 Recurrent or prolonged inflammation of the peritoneum results in the loss peritoneal space and semipermeable property.49, 52, 83 Bacterial biofilm adhering on the PD catheter is an important cause of refractory peritonitis (defined as failure of the PD effluent to clear up after 5 days of appropriate antibiotics), relapsing episodes, as well as recurrent exit-site or tunnel infections.84, 85 Timely catheter removal is therefore crucial for the eradication of these infections. Published series suggest that fungal peritonitis episodes generally have a poor response to anti-fungal therapy without catheter removal.86, 87 After catheter removal, the effective antibiotics should be continued for at least two further weeks.11 It is often appropriate to consider returning to PD after catheter removal and a minimum of 2–3 weeks of temporary HD.

Guideline 16.1

During exit-site or tunnel infection, the exit site should be cleansed at least daily.88 Since exit-site and tunnel infections often lead to subsequent peritonitis, they should be promptly identified and aggressively treated. The most serious and common pathogens are Staphylococcus aureus and Pseudomonas aeruginosa, which could be treated effectively on an out-patient basis with oral penicillinase-resistant (or broad spectrum) penicillin and fluoroquinolone, respectively.88, 89 Most ESI should be treated with at least 2 weeks of effective antibiotics, but ESI caused by Pseudomonas species or tunnel infection should be treated with at least 3 weeks.88 Refractory exit-site or tunnel infection is defined as failure to respond after 3 weeks of effective antibiotic therapy; simultaneous removal and re-insertion of the dialysis catheter with a new exit site under antibiotic coverage should be considered.88 A number of other interventions have been tried for the treatment of chronic or refractory catheter infections, but their evidence is limited.

Guideline 17.1

The overall benefit of prophylactic perioperative intravenous antibiotics is supported multiple randomized control trials.29 Although first-generation cephalosporin may be slightly less effective than vancomycin, the former is still commonly used because of the concern regarding vancomycin resistance. Current evidence does not support the use of any specific insertion technique, catheter design or PD solution for the prevention of peritonitis or catheter-related infections. Disconnect PD systems with a ‘flush before fill’ design results in a lower peritonitis rate as compared with the traditional spike systems,11 but spike systems are no longer routinely available for clinical use.

Guideline 17.2

Staphylococcus aureus peritonitis and ESI are associated with nasal carriage.90 Although there is no good randomized study to support routine screening of nasal S. aureus carriage in PD patients, this practice is easy to execute and facilitate the infection control procedures within hospital. The efficacy of prophylactic intra-nasal antibiotics, especially intra-nasal mupirocin, for the treatment of nasal carriage of S. aureus has been shown to reduce the risk of catheter-related infections, but the effect on peritonitis rate is less clear.29, 91, 92 Cyclical oral rifampicin therapy (typically 5 days course every 3 months) is also effective in reducing the rate of catheter-related infections, but the routine use of oral rifampicin for prophylactic purpose should not be recommended because adverse reactions, drug interaction and rifampicin resistance are all important problems. There are also good data to support the use of PD catheter exit-site antibiotic cream (either mupirocin or gentamicin) in all patients,30, 91, 93, 94 but emergence of resistant organisms is a concern when the use of topical antibiotic is prolonged and across the board.95, 96

Guideline 17.3

Invasive interventional procedures (e.g. colonoscopy, hysteroscopy, cholecystectomy) frequently cause peritonitis in PD patients.97 Several studies confirm that intravenous antibiotic prophylaxis before invasive gastrointestinal (GI) procedures reduces early peritonitis in these patients.29 The optimal antibiotic regimen is not well defined, but previous systematic review recommended the use of intravenous ampicillin plus an aminoglycoside, with or without metronidazole, for this purpose.29

Guideline 17.4

Most of the fungal peritonitis episodes are preceded by courses of antibiotics, often for the treatment of bacterial peritonitis episodes.98 A number of observational studies and randomized trials have shown that oral nystatin significantly reduces the risk of secondary fungal peritonitis following antibiotic therapy.29, 99, 100 Oral fluconazole is also effective for this indication,101 but there are potential problems (e.g. drug interactions, emergence of resistant strains) with fluconazole prophylaxis so that it is not routinely recommended.

Guideline 18.1

Acute PD has many potential advantages over extracorporeal dialysis for temporary renal support.103 Data on acute PD in AKI are summarized in a recent systematic review.104 After reviewing 24 studies with over 1500 patients, it was concluded that there is currently no evidence to suggest significant differences in mortality between acute PD and extracorporeal blood purification in AKI.104 As compared with other blood-based extracorporeal renal support, acute PD should particularly be considered in patients who are hypotensive, have contraindications to systemic anticoagulation, or have limited vascular access.

Guideline 18.2

There is little difference in outcomes between various methods of acute PD catheter insertion.105, 106 The ISPD guidelines on peritoneal access recommend that the method of insertion should depend on expertise at the centre.27 To ensure timely dialysis in the emergency setting, the current ISPD guideline emphasizes the need of, with the appropriate training, allowing nephrologists to insert acute PD catheters.102