Peritoneal Dialysis Watch — Sep 11, 2026

Welcome to your bi-weekly update in Peritoneal Dialysis.

This period is dominated by infection — how to prevent it, how to predict it, and what an exit-site infection actually tells you about what's coming next. Alongside that, there's a cluster of work on the prescription itself: glucose-sparing solutions, icodextrin for the long dwell, and a provocative cohort suggesting that in incremental PD, less dietary protein may be perfectly safe. And at the edges, evidence on heart failure therapy, walking capacity, and whether patients with myeloma-related kidney disease belong on PD at all. If you run a PD program, the infection material alone justifies the next nine minutes.

Let's start with infection prevention, because the American Society of Nephrology has now put a stake in the ground. In the Clinical Journal of the American Society of Nephrology, Perl and colleagues present a set of core interventions for preventing PD-related infections across United States dialysis facilities. The clinical question is structural rather than biological: infection rates vary enormously between centers, and the workgroup asked what a facility must do consistently to close that gap. Their answer rests on six pillars — routine surveillance with feedback of infection rates, standardized staff training with competency assessment, standardized patient and care-partner education, scheduled infection prevention assessments, antimicrobial prophylaxis at the catheter exit site, and prophylactic antimicrobials around specific procedures and events. The finding, if you can call it that, is a consensus judgment: the evidence base for each element is uneven, but the variation in outcomes between facilities is not explained by patient mix — it's explained by process. This is expert consensus built through iterative review, not a trial, and the authors are explicit that several recommendations rest on international guidelines rather than new data. What to do differently is concrete: if your unit is not feeding infection rates back to the staff who generate them, and not reassessing patient technique on a schedule rather than on suspicion, you have a gap you can close this quarter.

That guideline gains real force from a cohort in Frontiers in Public Health. Nochaiwong and colleagues asked what happens after an exit-site infection, using a multicenter Thai registry of more than five thousand incident PD patients, from which they matched three hundred and five patients with a first exit-site infection to three hundred and five without. The finding is a steep and time-dependent hazard: having an exit-site infection raised the subsequent risk of peritonitis roughly five-fold overall, and in landmark analysis the risk peaked at around eighteen-fold within the first month, decaying to about double by nine to twelve months. That pattern held regardless of causative organism or infection severity, and an accompanying meta-analysis of three cohorts pointed the same direction, at roughly threefold. It's retrospective and observational, matching cannot exclude that the patients who get exit-site infections are simply the patients with worse technique or worse hygiene, and the cohort was almost entirely continuous ambulatory PD in a single country. But the clinical message is hard to argue with. An exit-site infection is not a local event you treat and close out — it is a thirty-day window of extreme peritonitis risk, and it should trigger technique re-assessment, close follow-up, and a lower threshold for revisiting catheter management.

The complement to that is prediction before anything has happened, which is what Li and colleagues attempt in PLOS One. Across more than two thousand PD patients from ten centers, they compared six inflammatory indices derived from the routine complete blood count against subsequent peritonitis. Around seven percent of patients developed peritonitis over follow-up, and three of the indices — the pan-immune inflammatory value, the systemic immune-inflammation index, and the platelet-to-lymphocyte ratio — each roughly doubled the risk when elevated at PD initiation, consistently across subgroups. The honest finding, though, is that discrimination was only moderate; these are stratification tools, not decision rules. The study is retrospective, the cut-offs were derived in the same data they were tested in without external validation, and no index outperformed the others meaningfully. So the practical use is modest: a markedly elevated inflammatory ratio at initiation is a reason to invest more education and closer surveillance in that patient, not a reason to change the prescription.

And education is where the loop closes. In the Journal of Nephrology, Tan and colleagues pooled eleven randomized trials with just under a thousand PD patients testing structured educational interventions. Knowledge improved substantially, quality of life improved modestly — about five points on the scales used — and self-management and complication rates moved in the right direction. The trials were small, heterogeneous in content and delivery, at meaningful risk of bias, and GRADE certainty was limited, so the effect size on complications should be held loosely. But this is the evidentiary underpinning for the education pillar in the guideline, and it supports treating patient and caregiver retraining as a recurring intervention rather than a one-time onboarding event.

Turning to the prescription itself, where three papers converge on glucose, nutrition and the long dwell — and where they don't entirely agree. In BMC Nephrology, Lindholm and colleagues offer a systematic review of amino acid-based PD solutions, making the case that substituting one daily glucose exchange with a one-point-one percent amino acid solution preserves ultrafiltration and solute clearance while cutting cumulative intraperitoneal glucose load and supplying amino acids to malnourished patients. The evidence they synthesize is genuinely mixed in quality — randomized trials plus mechanistic work, with membrane and metabolic surrogates rather than hard patient outcomes, and no pooled effect estimate. Read it as a rationale for targeted use in high-risk or malnourished patients, not as a population-wide switch.

Running parallel is a prospective study in the Journal of International Medical Research, where Xie and colleagues took fifty continuous ambulatory PD patients through six months of standard glucose solution followed by six months with icodextrin for the overnight dwell. Ultrafiltration, Kt/V, albumin and hemoglobin all rose; C-reactive protein and B-type natriuretic peptide fell. That's a coherent picture of better volume control and less inflammation — but it is single-arm and sequential, so secular trends, intensified clinical attention, and regression to the mean are all uncontrolled, and the ordering means every patient received icodextrin second. It supports icodextrin as an individualization tool in patients with inadequate ultrafiltration, which most of us already believe; it does not establish the magnitude of benefit.

Now the tension. In Nutrients, Borrelli and colleagues report a one-year multicenter pragmatic cohort of two hundred and five patients starting incremental PD, comparing prescribed protein intake at or below zero point six grams per kilogram per day against higher intakes. Just under half of patients were on the lower-protein prescription, and after a year, serum albumin trajectories, body mass index, and persistence on incremental PD were essentially the same in both groups — while urea, phosphate and potassium were all better controlled on the lower-protein diet. That sits awkwardly beside the amino acid solution literature, which is premised on PD patients needing more protein and amino acid delivery, not less. Both cannot be the default. The resolution probably lies in residual kidney function — incremental PD patients with preserved clearance may simply be a different nutritional population from anuric long-vintage patients — but this cohort was not randomized, prescription was by local protocol, adherence was not measured against prescription, and albumin plus body mass index are weak proxies for lean mass. What would settle it is a randomized trial of protein targets in incremental PD with body composition and hard outcomes, not biochemistry. Until then, don't generalize a low-protein prescription beyond patients with real residual function.

Two papers look past the peritoneum to the rest of the patient. In the Jornal Brasileiro de Nefrologia, Silva and colleagues pooled nine studies and three hundred and forty-three PD patients on sacubitril-valsartan, finding ejection fraction improved by about five points, NT-proBNP fell substantially, and systolic pressure dropped roughly fifteen millimeters of mercury, with hyperkalemia in about one in eight patients and angioedema and hypotension rare. The critical caveat is that this was a single-arm meta-analysis with only one randomized trial in it, so before-and-after change is entangled with everything else that happens when a heart failure patient enters a study. It's permission to use the drug with potassium vigilance, not proof of benefit. Alongside that, in Therapeutic Apheresis and Dialysis, Sonoda and colleagues randomized twenty-four PD patients with sarcopenia or musculoskeletal pain to Nordic walking or usual care for twelve weeks. Six-minute walk distance improved by about thirty meters in the intervention arm versus almost nothing with usual care — borderline statistically, and in twenty-four patients that is a signal, not a conclusion. What is notable is ninety-six percent adherence in patients averaging seventy-four years of age, which tells you the intervention is feasible and worth offering while we wait for a properly powered trial.

Finally, a question of who gets offered PD at all. In Peritoneal Dialysis International, Jolec and colleagues used the French Language Peritoneal Dialysis Registry to find ninety-six patients with kidney disease from multiple myeloma among nearly thirteen thousand PD patients, matched four-to-one against other causes. Despite heavier comorbidity, myeloma was not associated with higher mortality or more transfer to hemodialysis — but these patients had roughly a fifth the chance of reaching transplantation. With fewer than a hundred cases, confounding by indication is unavoidable and survivors selected onto PD may be the healthier myeloma patients. Still, the actionable finding is that myeloma is not a reason to steer someone away from PD, and the transplant gap deserves scrutiny.

If you read only one paper from this period, make it the exit-site infection cohort in Frontiers in Public Health. It converts a familiar nuisance diagnosis into a quantified, time-limited high-risk window, and it tells you exactly when to intervene.

Here is what this period adds up to. First, infection prevention has moved from local custom to a defined process standard, and surveillance with feedback plus recurring retraining is now the expectation rather than a nicety. Second, exit-site infection should be treated as a peritonitis precursor with its highest hazard in the first thirty days. Third, risk stratification from routine blood counts is real but modest — useful for allocating attention, not for making decisions. Fourth, glucose-sparing strategies are supported by rationale and surrogates rather than outcomes, and the right protein target in incremental PD is genuinely unsettled. And what to watch for: adequately powered randomized trials of sacubitril-valsartan and of exercise in PD, both currently resting on evidence too thin for confident practice change.

That's your Peritoneal Dialysis update for this period. Until next time.

This is an automated summary generated by artificial intelligence, which can make mistakes. Always review the original source materials.