Noninfectious and Metabolic Complications

Applied Nephrology Master Series

Chapter 13

Noninfectious and Metabolic Complications

Abnormal Effluent | Hemoperitoneum | Eosinophilia | Chylous Effluent | Glucose Load | Icodextrin Safety | Hypokalemia | Surveillance

0. One-page chapter map

Table 13.1 — The eight decisions that govern noninfectious and metabolic-complication care.

Learning outcomes

• Distinguish noninfectious effluent abnormalities from PD-associated peritonitis without delaying appropriate infectious evaluation.

• Use effluent colour, cellular differential, timing and clinical severity to classify hemoperitoneum, eosinophilic inflammation and chylous effluent.

• Recognise when bloody effluent is compatible with a benign gynecologic phenomenon and when it signals intra-abdominal hemorrhage or surgical disease.

• Apply the conventional literature definition of eosinophilic peritonitis cautiously and understand that infection can coexist with peritoneal eosinophilia.

• Approach milky effluent through triglyceride/lymphatic and medication pathways rather than treating appearance alone.

• Explain why glucose-containing PD solution functions as both an osmotic prescription and a systemic metabolic exposure.

• Interpret HbA1c cautiously in dialysis and use directly measured glucose or CGM-derived information when HbA1c and clinical glucose patterns disagree.

• Use icodextrin and other glucose-sparing strategies as targeted prescription tools while protecting sodium-volume control and observing product-specific safety warnings.

• Recognise hypokalemia as a common PD vulnerability signal, identify reversible causes and distinguish trial targets from guideline recommendations.

• Interpret hypoalbuminemia and peritoneal protein loss without reducing nutrition assessment to a single serum albumin value.

• Cross-reference Chapter 14 for long-term membrane failure/EPS and Chapter 15 for detailed protein-energy wasting and nutrition management.

1. Core concept: abnormal effluent and metabolic drift are different PD signals

Noninfectious PD complications in this chapter fall into two clinically distinct groups. The first becomes visible in the drained bag: blood, unusual cloudiness, milkiness, fibrin or another colour change. The second evolves in the patient: hyperglycemia, glycemic variability, weight/adiposity change, dyslipidemia, hypokalemia and the consequences of protein loss. Both are PD-specific clues, but they should not be solved with the same reflex. [1,2,8]

Table 13.2 — Two-signal model.

2. The safety screen: benign mimics are diagnoses of exclusion when the phenotype is dangerous

The chapter starts where Chapter 10 left off: abnormal effluent can be noninfectious, but infectious peritonitis remains the first safety concern when the fluid is cloudy or the patient has abdominal pain. Severe focal pain, guarding, hemodynamic instability, rapid hemoglobin fall, pregnancy-related concern, recent abdominal procedure, bile-stained fluid or systemic toxicity move the patient out of a routine “effluent mimic” pathway. [2,8–10]

Table 13.3 — Red flags that bypass routine noninfectious troubleshooting.

3. Abnormal effluent: colour narrows the differential, cellularity changes the probability

A transparent bag is the normal visual baseline. When appearance changes, ask four questions: what colour is it, is it truly turbid or simply pigmented, what cells dominate, and what happened immediately before the change? Effluent WBC/differential and culture are central whenever peritonitis remains plausible; RBC count, triglyceride/chylomicron testing or targeted chemistry is added according to phenotype. [2,8]

Table 13.4 — Effluent phenotype to first diagnostic move.

4. Hemoperitoneum: dramatic colour does not equal severe bleeding, but severity must be proved

PD provides a “window” into the peritoneal cavity, so very small quantities of blood can colour a large volume of dialysate. Historical series show that many episodes are benign, particularly menstruation/ovulation-related bleeding, yet the same presentation can arise from gynecologic emergencies, recent procedures, trauma, cyst rupture, solid-organ or gastrointestinal disease, malignancy and long-vintage peritoneal pathology. [9,10]

Table 13.5 — Hemoperitoneum: phenotype to action.

5. Eosinophilic peritonitis: an inflammatory mimic, not permission to ignore infection

Eosinophilic peritonitis is a noninfectious inflammatory phenotype most often described early after catheter insertion or PD initiation and has been linked to exposure to air, catheter/solution materials, drugs and hypersensitivity-type reactions. A commonly used literature definition is >100 eosinophils/mL in effluent with eosinophils >10% of total effluent WBC; this is a conventional research/clinical definition rather than a new ISPD diagnostic criterion. [11,12]

Crucially, peritoneal eosinophilia can occur during infectious peritonitis. Therefore the presence of eosinophils does not safely cancel the Chapter 10 infectious pathway when symptoms, neutrophilia or culture findings support infection. [2,11,12]

Table 13.6 — Eosinophilic phenotype.

6. Chemical and sterile peritoneal inflammation: reconstruct the exposure

Chemical/sterile peritoneal inflammation is uncommon and is usually inferred from a temporal relationship to a dialysate, disinfectant, medication, connector contamination, solution lot or another intraperitoneal exposure after infectious causes have been appropriately investigated. The syndrome can resemble peritonitis clinically and cytologically; the key clinical task is exposure reconstruction rather than simply assigning the label “culture negative.” [2,8]

Table 13.7 — Exposure reconstruction.

7. Milky or white effluent: identify lipid/lymph rather than guessing from appearance

Milky effluent can reflect chyloperitoneum—triglyceride-rich lymph within the peritoneal cavity—but visual appearance alone cannot separate chylous fluid from infection, cholesterol-rich pseudochylous fluid or lymph-rich variants. General chylous-ascites literature often uses elevated fluid triglyceride and/or chylomicrons to establish the diagnosis; PD-specific interpretation should account for dialysate dilution and local laboratory methods rather than imposing one universal cutoff. [8,13]

PD-specific causes include catheter/surgical lymphatic injury, malignancy or lymphatic obstruction, pancreatitis and medication-associated chylous effluent. Dihydropyridine calcium-channel blockers have a well-described reversible association in PD case literature and systematic review. [14]

Table 13.8 — Milky effluent reasoning.

8. Fibrin and unusual colour: a clue to inflammation or another source, not a diagnosis

Fibrin strands may occur after catheter insertion, bleeding, peritonitis or other peritoneal inflammation and can contribute to catheter obstruction. Their management therefore belongs partly to the mechanical-access pathway in Chapter 12. The diagnostic mistake is to interpret fibrin as proof of infection—or to focus on fibrin clearance while missing the inflammatory source. [2]

Unusual orange, green, yellow, purple or black effluent has been reported with blood breakdown, medications, rhabdomyolysis, bile and other pigments. These are low-frequency phenotypes in which history and targeted testing dominate; green/bilious or feculent appearances should lower the threshold for urgent intra-abdominal evaluation. [8]

9. Glucose absorption: the osmotic prescription becomes a systemic metabolic exposure

Glucose is an effective and familiar crystalloid osmotic agent, but a clinically meaningful fraction is absorbed across the peritoneum. The absorbed amount varies with glucose concentration, dwell duration, transport characteristics and prescription architecture. The consequences can include higher glycemic exposure, insulin requirements, weight/adiposity gain and a triglyceride-rich lipid pattern, although the strength of evidence linking glucose absorption itself to hard cardiovascular outcomes remains incomplete. [1,4–6,16–18]

Table 13.9 — Glucose exposure: benefit, cost and safeguard.

10. Dysglycemia: measure the patient’s glucose exposure, not just HbA1c

Diabetes management in PD is complicated by glucose absorption from dialysate and by the reduced reliability of HbA1c in advanced CKD and dialysis. KDIGO 2022 states that HbA1c remains a monitoring tool but its accuracy and precision decline in dialysis; CGM-derived glucose management information can be used when HbA1c does not match directly measured glucose or clinical symptoms. [3,4]

Prescription changes can alter glucose exposure rapidly. A patient who shifts from repeated dextrose to a glucose-sparing regimen may need active review of insulin or other glucose-lowering therapy to avoid hypoglycemia, while increased hypertonic dextrose use during volume stress can raise glucose requirements. Diabetes-specific drug selection and targets are deferred to Chapter 16 and current diabetes/dialysis guidance. [3,4]

Table 13.10 — Glycemic assessment in PD.

11. Glucose-sparing prescriptions: metabolic improvement can create a volume problem

The IMPENDIA/EDEN randomized program showed that a glucose-sparing regimen using icodextrin and amino-acid solution improved HbA1c and several lipid measures in patients with diabetes, but the intervention group had more serious adverse events and deaths, including events compatible with extracellular fluid expansion. The lesson is not that glucose sparing is unsafe; it is that metabolic endpoints cannot outrank volume control. [5]

Table 13.11 — Glucose-sparing decision.

12. Icodextrin safety: the glucose-meter interaction is a never-event risk

Icodextrin is metabolised to oligosaccharides including maltose. Current product information warns that GDH-PQQ, glucose-dye-oxidoreductase and some GDH-FAD glucose-monitoring methods can produce falsely elevated glucose readings. This has caused inappropriate insulin administration or failure to recognise hypoglycemia. Falsely elevated readings may persist for up to approximately 2 weeks after stopping icodextrin. [15]

Compatibility of continuous glucose monitors and other devices is product-specific and should be checked with the current device manufacturer rather than assumed. This safety information must travel with the patient across emergency departments, hospital wards and procedures. [15]

Table 13.12 — Icodextrin safety checklist.

13. Weight gain: separate fluid, lean tissue and adiposity before changing the prescription

Weight gain after PD initiation is not synonymous with glucose-induced fat gain. Fluid status, return of appetite after uremia, changes in lean mass, reduced activity, insulin exposure and absorbed dialysate glucose can all contribute. The clinical task is to separate extracellular fluid expansion from tissue gain because the treatments are opposite. [1,17]

Table 13.13 — Weight gain differential.

14. Dyslipidemia: common metabolic phenotype, uncertain PD-specific outcome target

PD commonly produces or amplifies hypertriglyceridemia, VLDL/remnant abnormalities and other atherogenic lipid changes through insulin resistance, glucose absorption and kidney-failure metabolism. Contemporary reviews also explore possible links between dyslipidemia and peritoneal membrane outcomes, but causality and PD-specific treatment targets remain uncertain. [16–18]

The practical response is to avoid assuming that stronger glucose-based dialysis is metabolically neutral, assess cardiovascular risk in the wider CKD/dialysis context, and follow current lipid guidance rather than inventing a PD-specific lipid target. Changes in glucose exposure may modestly improve triglyceride/VLDL indices, but hard-outcome benefit from glucose sparing has not been established. [5,16]

15. Hypokalemia: a common PD vulnerability signal

ISPD 2020 goal-directed recommendations identify hypokalemia as associated with poor nutritional intake and adverse outcomes including peritonitis, and state that dietary and/or oral potassium supplementation should be considered. The 2024 systematic review/meta-analysis found hypokalemia common and associated with higher all-cause mortality, cardiovascular mortality and peritonitis, but certainty was low to very low and causality cannot be assumed. [1,6]

A 2022 multicentre randomized trial enrolled PD patients with recurrent/time-averaged potassium <3.5 mEq/L and compared protocol-based oral potassium supplementation targeting 4–5 mEq/L with reactive supplementation below 3.5 mEq/L. Protocol treatment increased serum potassium and reduced first-peritonitis hazard, but the trial was open-label and not designed to establish 4–5 mEq/L as a universal ISPD target. Hyperkalemia occurred in a small number of protocol-treated patients. [7]

Table 13.14 — Hypokalemia: cause before replacement.

16. Protein loss and hypoalbuminemia: albumin is a risk signal, not a nutrition diagnosis

PD causes ongoing peritoneal protein losses, which can increase during peritonitis and coexist with poor intake, inflammation, volume expansion and comorbidity. ISPD notes that hypoalbuminemia is more common in PD than HD and is associated with protein-energy wasting and peritoneal protein loss. [1]

Serum albumin is therefore clinically important but nonspecific. A low value should trigger assessment of inflammation/infection, intake, gastrointestinal disease, liver disease, urinary loss when relevant, volume state and dialysis-related protein loss. Detailed dietary protein prescription, PEW diagnosis, body-composition assessment and amino-acid solution use belong to Chapter 15. [1]

Table 13.15 — Low albumin: do not collapse the differential.

17. Integrated surveillance: make metabolic burden visible without duplicating later chapters

A high-quality PD review should make noninfectious and metabolic complications visible before they become technique-threatening. The minimum useful pattern is longitudinal: ask about abnormal effluent, review glucose-containing solution use and rescue hypertonicity, inspect glucose data in patients with diabetes, separate weight from volume, trend potassium, lipids and albumin in context, and link every abnormality to a named mechanism and reassessment plan. [1,3,4]

Table 13.16 — Integrated review dashboard.

18. Major clinical algorithms

19. Retention tables: pattern recognition

Table 13.17 — If you see this, think this first.

Table 13.18 — What not to confuse.

20. Clinical pearls

• 1. The bag is a diagnostic window: colour gets attention; cells and clinical context make the diagnosis.

• 2. Cloudy effluent remains infectious until appropriately evaluated—even when a noninfectious mimic is plausible.

• 3. A tiny amount of intraperitoneal blood can make a dramatic red bag; severity comes from hemodynamics, hemoglobin and source.

• 4. Menstruation/ovulation is common in hemoperitoneum, but persistent or severe bleeding deserves a broader intra-abdominal search.

• 5. Eosinophilic peritonitis is often early and benign, but eosinophils do not exclude infection.

• 6. A “culture-negative peritonitis” that stays eosinophil-predominant should trigger an exposure and differential-count review.

• 7. Milky effluent can be medication-related; always review recent calcium-channel blocker changes.

• 8. Fibrin is an inflammatory/access clue, not a stand-alone infectious diagnosis.

• 9. Dialysate glucose is a metabolic dose as well as an osmotic prescription.

• 10. The correct glucose-sparing prescription is the one that preserves fluid and sodium control.

• 11. HbA1c is less reliable in dialysis; discordant direct glucose data deserve more weight.

• 12. Icodextrin glucose-meter incompatibility is a high-consequence medication-device safety issue.

• 13. Weight change in PD must be split into fluid, fat, lean tissue and nutritional recovery.

• 14. Persistent hypokalemia is frequently a marker of poor intake or illness—not merely dialysate potassium loss.

• 15. Serum albumin is a prognostic risk marker, not a one-test diagnosis of malnutrition.

21. Common pitfalls — and the correction

Table 13.19 — High-frequency errors.

22. Mini-cases: decisions, not trivia

Case 1 — Pink effluent on a predictable cycle

A stable premenopausal patient has painless pink dialysate around mid-cycle, normal BP, no hemoglobin fall and spontaneous clearing.

Case 2 — Red bag with pain and dizziness

A patient presents with bright-red effluent, new focal abdominal pain, dizziness and a falling hemoglobin after an abdominal procedure.

Case 3 — Cloudy fluid, eosinophils, negative cultures

Two weeks after PD initiation, the effluent becomes cloudy. The patient is well, cultures remain negative and the differential is strongly eosinophilic rather than neutrophilic.

Case 4 — Milky effluent after a medication change

A patient develops painless milky dialysate shortly after a new dihydropyridine calcium-channel blocker. Cell count is low and cultures are negative.

Case 5 — Better HbA1c, worse edema

A diabetic APD patient is moved to a glucose-sparing regimen. Glucose indices improve, but weight, BP and edema increase.

Case 6 — Icodextrin and an unexpectedly high bedside glucose

An emergency department gives insulin for a very high point-of-care glucose in a patient using icodextrin, but the patient is confused and diaphoretic.

Case 7 — Persistent potassium 3.2–3.4 mmol/L

A PD patient repeatedly has low potassium, poor appetite and recent diarrhea. There is no acute ECG instability.

Case 8 — Low albumin after peritonitis

Albumin falls after a severe peritonitis episode while the patient remains edematous and anorexic.

23. Active recall

MUST MEMORIZE

Table 13.20 — Core facts.

USE AS REFERENCE

• Exact acute potassium-replacement doses and ECG-monitoring thresholds.

• Patient-specific glycemic targets and dialysis-compatible diabetes drug selection.

• Current glucose-meter/CGM compatibility with icodextrin.

• Laboratory-specific chylous-effluent triglyceride/chylomicron interpretation.

• Investigation of recurrent hemoperitoneum according to gynecologic, surgical and malignancy context.

• Solution- or lot-specific investigation of suspected chemical peritonitis.

24. Flashcards: spaced repetition

1. Q: Cloudy bag: first assumption? A: Infectious peritonitis until appropriately evaluated.

2. Q: Red bag: first severity question? A: Is the patient unstable, in severe pain, or losing hemoglobin?

3. Q: Common benign cause of hemoperitoneum? A: Menstruation/ovulation-related bleeding.

4. Q: Eosinophilic peritonitis typical timing? A: Often early after catheter insertion/PD exposure.

5. Q: Can infection cause peritoneal eosinophilia? A: Yes.

6. Q: Milky effluent: first lab direction? A: Cell count/culture if indicated plus triglyceride/chylous assessment.

7. Q: CCB class linked to chyloperitoneum? A: Dihydropyridine calcium-channel blockers.

8. Q: Fibrin means infection? A: No; it is an inflammation/bleeding/access clue.

9. Q: Why can PD worsen glycemia? A: Systemic absorption of dialysate glucose.

10. Q: HbA1c reliability in dialysis? A: Reduced compared with earlier CKD; use direct glucose/CGM if discordant.

11. Q: Glucose-sparing trade-off to watch? A: Extracellular volume expansion/insufficient UF.

12. Q: Icodextrin safety phrase? A: Use a compatible glucose-specific measurement method.

13. Q: Why can incompatible meters read high? A: Maltose/oligosaccharides interfere with some enzymatic methods.

14. Q: How long after stopping icodextrin can interference persist? A: Up to about 2 weeks per current product information.

15. Q: Weight gain first split? A: Fluid versus tissue.

16. Q: Typical PD dyslipidemia pattern? A: Often triglyceride/VLDL-rich, though individual patterns vary.

17. Q: Persistent low potassium: common hidden driver? A: Poor intake/PEW or GI loss, often with medication/PD contributions.

18. Q: Does hypokalemia cause peritonitis? A: Association exists; causality is not proved.

19. Q: What did the potassium RCT show? A: Protocol supplementation raised potassium and reduced first-peritonitis hazard in selected patients.

20. Q: Universal potassium target from that RCT? A: No.

21. Q: Low albumin: why nonspecific? A: Inflammation, volume expansion, intake and peritoneal loss all affect it.

22. Q: Membrane failure / EPS chapter? A: Chapter 14.

23. Q: Nutrition / PEW chapter? A: Chapter 15.

24. Q: Final mental model? A: Safety screen → classify effluent/body signal → name mechanism → change one lever → verify.

25. Rapid differential / troubleshooting

Table 13.21 — Noninfectious/metabolic troubleshooting from problem to action.

26. Final revision sheet

Table 13.22 — One-minute revision.

TEN TAKE-HOME RULES

1) Cloudy effluent is infectious until appropriately assessed. 2) Red fluid is common but not automatically benign. 3) Eosinophils do not exclude infection. 4) Milky fluid needs lipid/lymph testing, not visual diagnosis. 5) Dialysate glucose is a systemic metabolic dose. 6) HbA1c has reduced reliability in dialysis. 7) Glucose sparing never outranks euvolemia. 8) Icodextrin requires glucose-meter compatibility. 9) Persistent hypokalemia requires a cause search; one RCT target is not a universal guideline target. 10) Low albumin triggers a whole-patient assessment and the Chapter 15 nutrition pathway.

Table 13.23 — One-minute bedside synthesis.

Rapid oral viva

• Explain why a culture-negative cloudy bag is not automatically a noninfectious complication.

• Classify hemoperitoneum into benign-pattern versus emergency-pattern presentations.

• Define eosinophilic peritonitis using the conventional literature definition and explain its limitation.

• Build the differential for milky PD effluent and describe the role of medication review.

• Explain why dialysate glucose must be treated as a systemic metabolic exposure.

• Explain when HbA1c becomes misleading in dialysis and how directly measured glucose can help.

• Describe the IMPENDIA/EDEN lesson about glucose sparing and extracellular volume.

• Explain the icodextrin glucose-meter interaction and how long the warning remains relevant after discontinuation.

• Build a cause-based approach to recurrent hypokalemia and explain why 4–5 mEq/L is not a universal PD target.

• Explain why serum albumin cannot diagnose PEW by itself.

27. Selected authoritative references

1. Brown EA, Blake PG, Boudville N, et al. International Society for Peritoneal Dialysis practice recommendations: Prescribing high-quality goal-directed peritoneal dialysis. Perit Dial Int. 2020;40(3):244–253. https://doi.org/10.1177/0896860819895364.

2. Li PKT, Chow KM, Cho Y, et al. ISPD peritonitis guideline recommendations: 2022 update on prevention and treatment. Perit Dial Int. 2022;42(2):110–153. https://doi.org/10.1177/08968608221080586. PMID: 35264029.

3. Kidney Disease: Improving Global Outcomes (KDIGO) Diabetes Work Group. KDIGO 2022 Clinical Practice Guideline for Diabetes Management in Chronic Kidney Disease. Kidney Int. 2022;102(5S):S1–S127.

4. Wijewickrama P, Williams J, Bain S, et al. Narrative Review of Glycemic Management in People With Diabetes on Peritoneal Dialysis. Kidney Int Rep. 2023;8(4):700–714. https://doi.org/10.1016/j.ekir.2023.01.040. PMID: 37069983.

5. Li PKT, Culleton BF, Ariza A, et al. Randomized, controlled trial of glucose-sparing peritoneal dialysis in diabetic patients. J Am Soc Nephrol. 2013;24(11):1889–1900. https://doi.org/10.1681/ASN.2012100987. PMID: 23949801.

6. Yang C, Hu X, Ling X, et al. Hypokalemia in Peritoneal Dialysis: A Systematic Review and Meta-analysis of Prevalence, Treatment, and Outcomes. Kidney Med. 2024;6(12):100923. https://doi.org/10.1016/j.xkme.2024.100923. PMID: 39634334.

7. Pichitporn W, Kanjanabuch T, Phannajit J, et al. Efficacy of Potassium Supplementation in Hypokalemic Patients Receiving Peritoneal Dialysis: A Randomized Controlled Trial. Am J Kidney Dis. 2022;80(5):580–588.e1. https://doi.org/10.1053/j.ajkd.2022.03.013. PMID: 35597332.

8. Dossin T, Goffin E. When the color of peritoneal dialysis effluent can be used as a diagnostic tool. Semin Dial. 2019;32(1):72–79. https://doi.org/10.1111/sdi.12740. PMID: 30032485.

9. Lew SQ. Hemoperitoneum: bloody peritoneal dialysate in ESRD patients receiving peritoneal dialysis. Perit Dial Int. 2007;27(3):226–233. PMID: 17468466.

10. Greenberg A, Bernardini J, Piraino BM, Johnston JR, Perlmutter JA. Hemoperitoneum complicating chronic peritoneal dialysis: single-center experience and literature review. Am J Kidney Dis. 1992;19(3):252–256. https://doi.org/10.1016/S0272-6386(13)80006-6. PMID: 1553970.

11. Zhang Q, Xia Y, Zhang M, Jiang C. Peritoneal dialysis related eosinophilic peritonitis: a case report and review of the literature. BMC Nephrol. 2023;24(1):10. https://doi.org/10.1186/s12882-022-03027-8. PMID: 36635670.

12. Uzun Kenan B, Büyükkaragöz B, Leventoğlu E, Bakkaloğlu SA. Eosinophilic peritonitis in children undergoing maintenance peritoneal dialysis: A case report and literature review. Semin Dial. 2022;35(6):548–555. https://doi.org/10.1111/sdi.13113. PMID: 35788998.

13. Bhardwaj R, Vaziri H, Gautam A, et al. Chylous Ascites: A Review of Pathogenesis, Diagnosis and Treatment. J Clin Transl Hepatol. 2018;6(1):105–113. https://doi.org/10.14218/JCTH.2017.00035. PMID: 29577037.

14. Chen YC, et al. Calcium Channel Blocker-Related Chylous Ascites: A Systematic Review and Meta-Analysis. Medicina (Kaunas). 2019;55(4):93. PMID: 30959848.

15. EXTRANEAL (icodextrin) peritoneal dialysis solution. Current U.S. prescribing information / DailyMed. Safety warning on glucose-monitor interference. Accessed 3 September 2026.

16. Stepanova N. Dyslipidemia in Peritoneal Dialysis: Implications for Peritoneal Membrane Function and Patient Outcomes. Biomedicines. 2024;12(10):2377. https://doi.org/10.3390/biomedicines12102377. PMID: 39457689.

17. Wang J, Zhao J, Li L, et al. Association Between Peritoneal Glucose Absorption, Lipid Metabolism, and Cardiovascular Disease Risk in Nondiabetic Patients on Peritoneal Dialysis. J Ren Nutr. 2025;35(1):196–206. https://doi.org/10.1053/j.jrn.2024.05.005. PMID: 38851308.

18. Holmes CJ. Adverse effects of systemic glucose absorption with peritoneal dialysis: how good is the evidence? Curr Opin Nephrol Hypertens. 2013;22(6):635–642. PMID: 24076559.

19. Morelle J, Stachowska-Pietka J, Öberg C, et al. ISPD recommendations for the evaluation of peritoneal membrane dysfunction in adults: classification, measurement, interpretation and rationale for intervention. Perit Dial Int. 2021;41(4):352–372. https://doi.org/10.1177/0896860820982218.