Lung Water by Ultrasound Guided Treatment in Hemodialysis Patients (The Lust Study).

Randomization procedure: After enrollment, patients will be randomized in 2 groups: treatment and control. The randomization will be performed by centre, with a 1:1 allocation ratio, and it will be communicated by the coordinating centre in Reggio Calabria. An allocation concealment will be observed.

Warning: based on the principle "once randomized, always randomized", if a patient drops out the study he cannot simply be replaced by a new one. New patients must be randomized as well.

Validation of core lab data: The core data to be collected in the LUST project/subprojects should be validated by the following validation centres:

• CORE LUST STUDY DATA Echocardiographic data and US-lung data: Dr. Rosa Sicari and Dr. Luna Gargani, CNR Institute of Clinical Physiology of Pisa, Italy (rosas@ifc.cnr.it).
• ANCILLARY LUST PROJECTS Pulse Wave Velocity data (PWV): Prof. Gerard London, Service d'Hémodialyse, Hôpital F.H. Manhès, Fleury-Mérogis, France (glondon@club-internet.fr). 24 Ambulatory Blood Pressure Monitoring (ABPM) data: Prof. Francesca Mallamaci, Nephrology, Dialysis and Transplantation Unit and IFC-CNR, Reggio Calabria

Preliminary training on US-B lines measurement is a pre-requisite for Nephrologists and Cardiologists of each participating centre. After training (Skype meeting) an online certification will be released to certify the level of expertise about US-B lines measurements of all LUST study nephrologists and cardiologists.

Each participant centre will upload pertinent studies (echocardiography, 24h ABPM and PWV data) into the LUST website and will timely receive a feedback by the validation centres, whenever needed.

US Lung Scan technique A standard (3.0-MHz) echocardiography probe can be used for this purpose. Examinations should be performed in the supine position. To detect lung water, both sides of the chest should be scanned both in the anterior and lateral part from the second to the fourth (on the right side to the fifth) intercostals spaces, at parasternal, mid clavea, anterio-axillar and mid-axillary lines, as previously described (Am J Cardiol 2004;93:1265-70). An US-B line is a hyperechoic, coherent US bundle at narrow basis going from the pleura to the limit of the screen. These extended lines (also called comets) arise from the pleural line and should be differentiated from short comets' artifacts that may exist in other regions. US-B lines starting from the pleural line can be either localized or scattered to the whole lung and be present either as isolated US-B lines or in groups (with a distance >7 mm between 2 extended comets). The sum of US-B lines produces a score reflecting the extent of LW accumulation. If you do not detect any US-B line, the score is 0.

The investigators warn the physicians participating in the trial about fixed US-B lines, due not to lung water but to fibrosis or inflammatory processes. These comets should be registered but not counted as US-B lines. The cause that underlies these artefacts should be described.

In patients allocated to the active arm of the study, any decision about weight reduction will be taken by Nephrologists only on the basis of pre-dialysis US-scans. However, also post-dialysis US-B lines measurement must be performed and registered. In the active and in the control arm as well, US-scans will be performed by the local cardiologist in coincidence with the echocardiographic study. It is crucial for the purpose of this study that cardiologists keep nephrologists involved in the LUST study blinded with respect to the number of US-B lines of patients of the control group.

Timing

• In patients allocated to the control and the active arms, US-B lines should be performed at baseline and at 6, 12 and 24 months by the local cardiologist, in a non-dialysis day either on a Thursday or on a Friday, depending on the dialysis schedule.
• In patients allocated in the treatment arm, the application of US-B scans should be performed as follows:
• In patients with less than 15 US-B lines, ultrafiltration (UF) will not be modified and the US-B lines monitoring will be performed by the nephrologist at monthly intervals;
• In patients with more than 15 US-B lines (either at baseline or at any of the monthly US-B Lung scans) UF will be performed by scheduling longer and/or additional dialyses (see below). In these patients, monitoring of US pre-dialysis and post-dialysis -B lines will be repeated at least once a week, until the goal is achieved (see next section).
• Furthermore, in patients in the active arm, pre- and post-dialysis US-lung scans can be repeated at discretion of the nephrologist, i.e. whenever he/she believes that it can be useful applying this technique for monitoring the volume status of the patient, for example in a patient with <15 US-B lines who subsequently manifests a rise in body weight or in a patient who develops hypotension or frank hypotensive episodes during dialysis .

In brief, in the treatment group, monitoring is scheduled once a week (before and after dialysis), or even more frequently if the nephrologist believes that a more frequent monitoring may be useful to track the desired UF goal. When the number of US-B lines falls below 15, the measurements will be repeated at monthly intervals.

Weight reduction in patients with more than 15 US-B lines

In patients randomized to the treatment arm and with a number of US-B lines >15, a decrease of dry weight is required to reduce lung water, according to the following scheme derived by a pilot study at the coordinating centre and at the Iasi Nephrology Unit:

• 15-30 US-B lines: decrease dry weight by 300 g over the following week (about 100 g per session)
• 31-40 US-B lines: decrease dry weight by 450 g over the following week (about 150 g per session)
• > 40 US-B lines: decrease dry weight by 600 g over next week (about 200 g per session).

Attempts to lower dry weight according to the previous scheme should continue until the US-B lines goal (<15) is attained. If the patient does not tolerate attempts to decrease dry weight for 2 weeks (i.e. if he/she develops hypotension, cramps and other symptoms) extra haemodialysis sessions should be considered.

If the goal (<15 pre-dialysis US B-lines) is not achieved after 4-6 weeks, or in case the patient does not tolerate UF, drug treatment (carvedilol, ACEi, ARB) intensification or introduction should be considered.

Once the goal is achieved, it is recommended to confirm it by repeating the measurement of US-B lines at least once, for example before the following dialysis.

Warnings:

• In patients in the active arm with initial pre-dialysis US-B lines <15 and in those who achieve this goal thanks to body fluids subtraction who develop cramps and/or symptoms of extracellular volume depletion (low dry weight) the decision of increasing dry weight should be accompanied by close US-B lines monitoring (i.e. every dialysis session until the patient stabilizes).
• In patients with >15 pre-dialysis US-B lines who do not tolerate dry weight decrease due to hypotension and who are on hypotensive/cardioprotective drugs treatment (carvedilol/ACEi) the dose of these drugs should be down-titrated and stopped, if needed. These drugs in these patients should be always given after dialysis (before going to bed, at night) rather than pre-dialysis.

Clinical (pre-dialysis) evaluation of volume status In both groups (treatment and control group), a standard pre-dialysis clinical evaluation of volume status should be done, as it is recommended in good clinical practice. This evaluation should consider blood pressure and change of blood pressure over time, pedal edema, presence/absence of dyspnea, crackles on lung auscultation (see below), body weight gain inter-dialysis and body weight trajectory over time. In the active arm of the study these data should also be formally registered whenever US-B lines measurements are done. In the control group, evaluation of volume status needs to be formally registered only in coincidence of the echocardiographic studies.

The following scale will be used for the evaluation of Crackles :

1. No crackles
2. I am uncertain about the presence of fine crackles
3. Definite fine crackles at lung bases
4. Moderate crackles
5. Bilateral, diffuse crackles

For clinical edema, the following scale will be used:

1. No clinical edema
2. Slight pitting (2 mm depth) with no visible distortion
3. Somewhat deeper pit (4 mm) with no readily detectable distortion
4. Noticeably deep pit (6 mm) with the dependent extremity full and swollen
5. Very deep pit (8 mm) with the dependent extremity grossly distorted

Echocardiographic measurements In both the intervention and control arm, cardiologists involved in the LUST study will perform the echocardiographic measurements at baseline, 6, 12 and 24 months, in a non-dialysis day on a Thursday or on a Friday, depending on the dialysis schedule. These echocardiogram are "research echocardiograms". For this reason, it is fundamental to keep nephrologists blind about the results of these echocardiographic readings.

Nevertheless, nephrologists are allowed to consult a cardiologist (be him the LUST cardiologist or another one) for a "clinical echocardiogram" if and when clinical problems arise which demand the application of echocardiographic studies for diagnostic reasons or for monitoring any underlying cardiac problem.

Detailed information about standardization of echocardiographic study has been prepared by echocardiography validation laboratory (Dr. Rosa Sicari and Dr. Luna Gargani, IFC CNR, Pisa).

Statistical analysis plan A total sample size of 500 patients (250 per group) will be expected to provide approximately 80% power to detect a difference in the primary end point with an assumed type I error rate of .05, 2-sided.Investigators estimate that the 2 year event rate for the composite end point would be 45% in the usual care group and 30% (a 33% risk reduction) in the arm with the lung US-guided intervention. According to protocol, all patients will be followed for 24 months after randomization.

Survival analysis will be performed by the Kaplan-Meier and the Cox regression methods. Missing baseline categorical variables will be replaced with the mean or median value, as appropriate. The effect of the allocation arm on the number of US-B lines will be investigated by Linear Mixed Models (LMM).

The between arms differences in echocardiographic data (secondary end point) will be compared by T-Test, Mann-Whitney test and Linear Model Analysis, as appropriate.

A secondary analysis of the longitudinal evolution of US B lines and the time to the first event and to repeated events will be performed by the Joint Models. All analyses will be based on the principle of intention to treat and will be performed using the SPSS version 24, STATA version 13. The threshold for statistical significance will be 2 sided with a type I error rate of .05.

Long-term effect of the intervention In order to capture any long-term effect of the study intervention, data on vital status and cardiovascular events occurred to all patients alive after the intervention period will be collected for additional 24 months after the end of the study.