Massive Dialysis Study Reveals Why Catheters Fail Mid-Treatment

For hundreds of thousands of people whose kidneys no longer work, haemodialysis is a lifeline that runs on a surprisingly fragile piece of plumbing: a tunnelled central venous catheter threaded into a large vein and used hundreds of times a year to shuttle blood through a dialysis machine. When that lifeline falters, the consequences cascade quickly. Blood flow drops, toxins linger in the body, dialysis sessions become less effective, and hospital visits multiply. Now one of the largest session-by-session analyses of this problem to date, drawing on more than 40,000 individual dialysis treatments across 15 Spanish centres, has mapped in unprecedented detail which clinical and technical factors travel alongside catheter dysfunction, offering clinicians a data-driven picture of when and why these devices fail.

The research, published in the open-access journal Nursing Open, took an unusual analytical approach: instead of studying patients as single units, the investigators examined every recorded haemodialysis session over a full year, from January to December 2021, in patients treated through tunnelled catheters within a national renal foundation network in Spain. The result was a dataset of 40,193 sessions from 743 patients, of which 987 sessions, or 2.5 percent, met the study’s operational definition of intradialytic catheter dysfunction. That definition was grounded in internationally recognised KDOQI thresholds: a machine-registered dysfunction flag appeared whenever blood flow fell below 300 millilitres per minute while arterial pre-pump pressure dropped below minus 250 millimetres of mercury, venous return pressure rose above plus 250 millimetres of mercury, or both, at the start of the session or at the 60-minute mark. Nursing staff then verified each flagged event in the electronic dialysis record system.

The cohort itself was clinically striking. The patients averaged 68.4 years of age and carried a heavy burden of chronic disease: 84.5 percent had arterial hypertension, 60.1 percent had cardiovascular disease, and nearly half lived with diabetes. Diabetic nephropathy was the leading cause of kidney failure, accounting for 34.3 percent of cases. Notably, women made up 63.7 percent of the cohort, a marked contrast with national registry data in which men typically account for roughly two-thirds of the haemodialysis population, a difference the authors attribute to the particular clinical profile of patients who end up dialysing through tunnelled catheters rather than through surgically created arteriovenous fistulas.

When the researchers compared dysfunctional sessions with smooth-running ones, the haemodynamic signatures diverged sharply, exactly as the outcome definition would predict. Sessions with dysfunction began with a mean blood flow of 257.9 millilitres per minute versus 350.7 in healthy sessions, and by the 60-minute mark the gap persisted, with flows of 256 versus 358.3. Arterial pressures plunged dramatically deeper into negative territory, averaging minus 215.2 millimetres of mercury across dysfunctional sessions compared with minus 172.2 in uneventful ones. Total blood volume processed fell from 75.6 litres to 56.2, and the final Kt, the routinely recorded marker of the dialysis dose actually delivered, dropped from 48.2 litres to 39.1. In plain terms, a failing catheter deprived patients of nearly a fifth of the intended cleansing capacity of each session.

Beyond the intradialytic pressure readings themselves, which the authors are careful to frame as manifestations of dysfunction rather than causes of it, the analysis identified baseline characteristics that clustered with failing sessions. Catheters placed in the left internal jugular vein showed a dysfunction rate of 5.1 percent, dramatically higher than the 0.7 percent observed with the right internal jugular vein, the most popular insertion site in the cohort. Femoral placement, though rare, performed poorly, with the right femoral vein showing dysfunction in 6.3 percent of sessions, consistent with earlier evidence that femoral and subclavian sites carry higher thrombotic and mechanical risks. Comorbidities also left fingerprints: diabetes, cardiopathy, peripheral arterial disease, prior stroke, and a history of malignancy were all significantly more frequent in dysfunctional sessions, as was systemic anticoagulant therapy, though the authors caution this may reflect the vascular risk profile of the patients rather than a drug effect.

Perhaps the most technically revealing part of the study was its stratification of dysfunction events into arterial and venous types. Of the 1,481 recorded events, 1,346 were arterial and 135 were venous, and the two behaved like distinct clinical entities. Arterial events featured severely negative pre-pump pressures, averaging minus 221.9 millimetres of mercury, signalling that the catheter was struggling to draw blood in, and were more common in older patients with hypertension, cerebrovascular disease, and prior SARS-CoV-2 infection. Venous events, by contrast, showed soaring return pressures averaging 220.4 millimetres of mercury, indicating obstruction on the outflow side, and occurred disproportionately in men, in patients with peripheral arterial disease, tumours, previous kidney transplants, and those on anticoagulants. Venous events also produced significantly lower final Kt values, suggesting the outflow type of failure may be particularly punishing for dialysis adequacy.

Two management variables, line reversal and 5 percent sodium heparin locking solution, appeared far more often in dysfunctional sessions, with line reversal documented in 3.5 percent of sessions without dysfunction versus an elevated share within dysfunctional ones, and 5 percent heparin locks showing a 4.3 percent dysfunction rate against 2 percent for the standard 1 percent heparin. But the researchers explicitly warn against reading these numbers causally. In routine practice, clinicians reverse catheter lines and escalate locking solutions precisely because a catheter is already underperforming, so the association plausibly reflects a rescue response rather than a culprit. This principle of reverse causality and confounding by indication permeates the entire analysis and is one of the paper’s most intellectually honest contributions to a field where such distinctions are often blurred.

The pharmacological findings followed the same cautious logic. Overall, systemic antiplatelet therapy showed no significant association with dysfunction when analysed as a single binary variable, yet drug-specific comparisons revealed significant heterogeneity, with acetylsalicylic acid regimens, clopidogrel, prasugrel, and dipyridamole each showing different dysfunction frequencies. Among anticoagulants, enoxaparin and especially tinzaparin were the agents most frequently associated with dysfunction events, while combined regimens such as enoxaparin with warfarin appeared far less often in failing sessions. The authors interpret these patterns as exploratory and hypothesis-generating, likely shaped by differences in patient indication and underlying vascular disease rather than by direct pharmacological effects on catheter patency, and they point to prior trials of agents like recombinant tissue plasminogen activator and taurolidine-based locks as avenues where randomised evidence has begun to clarify what observational data cannot.

Dialysis efficacy emerged as a second major storyline. Using sex-specific operational Kt thresholds, the researchers classified 24.7 percent of sessions as low efficacy, 20.8 percent as normal, and 52.6 percent as optimal. The sex split was dramatic: 41.4 percent of male patients’ sessions fell into the low-Kt category, while 65.6 percent of women’s sessions reached the optimal range, a highly significant difference partly attributable to the fact that Kt depends on body size, and men on average process more blood volume through larger bodies. Because Kt was measured session-by-session through ionic dialysance rather than sampled sporadically, the study demonstrates how continuous, per-session adequacy monitoring can flag patients at risk of chronic under-dialysis, particularly catheter-dependent men, long before cumulative toxin buildup becomes clinically apparent.

The study’s limitations are considerable and openly acknowledged. The analysis was univariate and did not adjust for the fact that dozens of sessions belonged to the same patient, which may have inflated statistical significance. Being retrospective and built on routine electronic records, it could not capture insertion technique, operator experience, catheter tip design or position, or fibrin sheath formation, all recognised determinants of catheter survival, and all data came from a single institutional network, limiting generalisability. Still, the practical message is clear and actionable: continuous vigilance over blood flows, line pressures, and pressure trajectories during every dialysis session can catch failing catheters early, and the distinct arterial-versus-venous signatures described here give nurses and nephrologists a framework for interpreting what the machine alarms are actually telling them. The authors call for adjusted analytical studies, trials of locking strategies and anticoagulation protocols, and exploration of nursing-led interventions, all aimed at turning a descriptive atlas of catheter failure into genuine prediction and prevention.

Subject of Research: Clinical and technical factors associated with tunnelled central venous catheter dysfunction in haemodialysis patients

Article Title: Clinical and Technical Factors Associated With Tunnelled Central Venous Catheter Dysfunction in Haemodialysis: A Multicentre Session‐Level Retrospective Study

Article References: Gimeno Hernán, V., Perez Ingidua, C., Rivas Paterna, A. B., Moreno, G., Vicente‐Galán, M. J., Orgaz‐Rivas, E., Meneses‐Monroy, A., & Ortuño‐Soriano, I. (2026). Clinical and Technical Factors Associated With Tunnelled Central Venous Catheter Dysfunction in Haemodialysis: A Multicentre Session‐Level Retrospective Study. Nursing Open, 13(9), Article e70758. https://doi.org/10.1002/nop2.70758

Image Credits: AI Generated

DOI: 10.1002/nop2.70758

Keywords: haemodialysis, central venous catheter, catheter dysfunction, dialysis adequacy, Kt, vascular access, anticoagulation, antiplatelet therapy, nephrology, internal jugular vein, retrospective study, nursing