Study setting {9}
This study will take place in six ICU/HAUs in hospitals located across Canada. A list of study sites can be found at clinicaltrials.gov under the ID: NCT04548713.
Eligibility criteria {10}
Inclusion Criteria:
-
ICU/HAU Admission
-
Presence of a central venous catheter requiring locking. This includes triple lumen central venous catheters (both tunneled and non-tunneled), dialysis lines, implanted vascular access devices (IVADs) and peripherally inserted central lines (PICCs).
-
>18 years of age
Exclusion Criteria:
-
Known sensitivity to EDTA
-
Confirmed or suspected pregnancy, as 4% tetrasodium EDTA has not yet been investigated for use in the pregnant population.
-
Patients who decline receiving blood products (due to the smaller blood draws and special blood conserving procedures).
-
Physician, patient, or temporary substitute decision maker (TSDM) declines
-
Currently enrolled in any other research study that may confound primary outcome measures. Co-enrolment in multiple studies will be considered on an individual basis.
-
Patients who were previously enrolled in the study. Patients who were enrolled in the first period are not eligible for (re-)enrolment in the second period, and patients who are enrolled in the study and transferred to another participating hospital are not eligible for (re-)enrollment at the receiving hospital. Patients who had been discharged from the unit to another hospital ward and are re-admitted to the critical care unit are no eligible for re-enrollment into the study.
Who will take informed consent? {26a}
Under the guidance of the Tri-Council Policy Statement 2, this study has been approved for waiver of consent (under Article 3.7A) by relevant ethics boards for the following reasons: 4% tetrasodium EDTA is considered at least as safe as standard of care and has Health Canada approval, there is a strong chance that 4% tetrasodium EDTA will be safer than standard of care, there have been no adverse outcomes reported in previous clinical use and the study could not be feasibly completed with a deferred or a full consenting process due to the rapid decision to use a CVAD lock and the need to use it immediately.
Additional consent provisions for collection and use of participant data and biological specimens {26b}
Not applicable.
Interventions
Explanation for the choice of comparators {6b}
The comparator to the trial intervention will be standard of care CVAD lock. For central venous catheters and peripherally inserted central catheters, this is saline. For hemodialysis catheters, citrate can be used. As the aim of the study is to determine whether 4% EDTA is more efficacious in preventing CVAD complications than the current standard of care, we must compare this alternative to current practices.
Intervention description {11a}
Upon admission to the ICU/HAU, any patient with a CVAD requiring locking who meets criteria will be enrolled into the study. The patient will receive either a standard of care lock or 4% tetrasodium EDTA locking solution depending on the randomization status of their treating hospital. Syringes containing either 4% tetrasodium EDTA solution or standard of care locking fluid to be used during locking procedures (called ‘locking kits’) will be prepared ahead of time by the research pharmacy according to randomization status of the unit. They will be clearly labelled for use as a locking solution only and placed at the patient’s bedside. Syringes will be prepared with an additional label to remind nurses not to use in pregnant patients or in those who decline any blood products. In addition to the bedside assessment, research coordinators will review each patient that has received a locking solution every day to gather data and to ensure compliance with study protocols. Nurses will use the prepared syringes to lock and maintain each CVAD according to standard of care. Syringe supply at each patient bedside will be replenished as needed by the research pharmacy. Enrolment can occur 24 hours per day, 7 days per week as nurses can obtain locking syringes specific to the randomization status of the unit at all times
Medication, maintenance fluids, pressors, or anesthesia will be administered through the CVAD according to standard of care. Once a lumen of the catheter is no longer being utilized for continuous infusions, leftover medication will be aspirated and flushed with 20mL of 0.9% sodium chloride using a turbulent flush method as per standard of care. For patients randomized into the experimental group, the catheter will then be locked with the appropriate volume of 4% tetrasodium EDTA, as specified by catheter manufacturer instructions. These procedures will remain exactly the same for patients randomized into the control group with the exception of using saline or citrate as a locking fluid in accordance with standard central venous catheter maintenance procedures. Experimental and control solutions are indistinguishable by visual inspection.
All further catheter procedures and maintenance will proceed according to standard of care. For all locked catheters, frequency of flushing will occur every 12 hours (or 72 hours for dialysis catheters). During this procedure, catheters locked with the standard of care lock will be aspirated, flushed with 10mL of 0.9% sodium chloride using turbulent flow, and re-filled with the appropriate volume of the standard of care lock. Procedures will stay exactly the same for patients in hospitals randomized to administer the 4% tetrasodium EDTA treatment, except for continued locking with 4% tetrasodium EDTA. Neutral displacement cap change procedures will remain unchanged in both groups. This includes a change every 96 hours, after a blood draw through a cap, if removed, contaminated, damaged, and as needed.
Regular assessments of the site of insertion, CVAD system, and patency, will continue according to standard of care, regardless of type of locking solution used. According to the Adult CVAD Maintenance Record used at each hospital, nurses will assess that dressing is secure, dry and intact, will palpate the site, and check the system at the beginning of each shift. The site of insertion and connections will be reassessed every four hours as is standard protocol. Dressing and tubing changes will also proceed according to standard of care in both groups.
Should a locked line (locked with either standard of care or 4% tetrasodium EDTA solution) need to be re-opened for additional medication administration, the line will be assessed for patency, aspirated of locking solution, and flushed with 20 mL of 0.9% sodium chloride before administering further medication. Patency will be assessed by the ability to aspirate for blood return and the ability to flush a CVAD without resistance prior to the administration of parenteral medications and solutions. If the line is not patent, the nurse will assess for occlusions and proceed with standard of care occlusion diagnosis and treatment. As per standard of care, central lines will continue to be removed at the earliest opportunity for each patient. 4% EDTA has been shown to be safe if it is inadvertently flushed into the patient, thereby ensure an adequate safety profile for use with these vulnerable patients.
Criteria for discontinuing or modifying allocated interventions {11b}
The patient may withdraw from the trial at any time. Patients for whom an exclusion criterion is discovered after enrollment will be withdrawn from the study and excluded from analysis.
Strategies to improve adherence to interventions {11c}
The patients themselves do not have to do anything to adhere to the intervention or provide measurements of primary and secondary outcomes. The intervention itself will be administered by ICU/HAU staff at each center. Strategies to ensure staff adhere to experimental procedures include education sessions on purposes and procedures of the trial before the start of the experiment and during the trial. ICU/HAU staff have also been a part of protocol development to ensure seamless incorporation of experimental procedures into standard workflow. The primary and secondary outcomes are already captured as part of standard of care and will be recorded in patient charts. These charts will be reviewed by research staff to fill out study Case Report Forms (CRFs).
Relevant concomitant care permitted or prohibited during the trial {11d}
Clinical care will continue as per discretion of the patient's healthcare team. The intervention is a CVAD locking fluid that remains only in the lumen of the CVAD, it will never actually enter the patient's circulation. The locking fluid will be aspirated out of the catheter following the end of locking. As such, there is no potential for drug interactions, and we do not foresee any restrictions in enrolling patients based on other types of concomitant care.
Provisions for post-trial care {30}
Not applicable.
Outcomes {12}
Primary Outcome:
-
Incidence rate of the following: CLABSI as diagnosed in accordance with the Fraser Health CLABSI diagnosis algorithm (see Additional File 1) based on NHSN guidelines, CVAD replacement due to occlusion, and/or catheter obstruction requiring Alteplase use. The composite incidence rate will be calculated as the number of events divided by the number of catheter-days. The patient is considered at risk during the time the patient is in the ICU/HAU and has a catheter placed. Multiple outcomes occurring within a 24-hour time frame will be counted as a single event, except for CLABSI, any occurrence less than 2 calendar days following a previous CLABSI and subsequent catheter removal will be considered attributed to the previous catheter, and treated as a single event.
Note about Primary Outcomes:
In the case of suspected CLABSI, diagnostic procedures will proceed according to the Fraser Health CLABSI Case Identification Algorithm, in all participating ICU/HAUs. Following CLABSI diagnosis, patients will receive antibiotic treatment as necessary. Should the care team decide that replacement of the CVAD be necessary, treatment will proceed according to standard of care. Upon subsequent locking procedures, patients will receive the same locking solution as the rest of the hospital unit at that particular point of study procedure. All cases of CLABSI will be adjudicated by a blinded, external infectious disease expert.
Secondary Outcomes:
-
Incidence rate of confirmed CLABSI (as adjudicated by an external expert according to the NHSN criteria described previously).
-
Incidence rate of suspected CLABSI, as adjudicated by an external infectious disease expert. A case of suspected CLABSI will be defined as a case in which no microbiological testing was performed on the suspected catheter, with the patient having a positive blood culture result of a common bacteria associated with a CVAD infection with no other identifiable source of infection.
-
Incidence rate of catheters requiring removal due to occlusion
-
Incidence rate of catheters requiring alteplase use
-
Incidence rate of catheter-associated venous thrombosis, as determined by inspecting relevant medical imaging and notes on patient charts.
-
Incidence rate of catheter colonization, as diagnosed according to the Clinical Practice guidelines of Infectious Diseases Society of America (30).
-
Classification of types of microorganisms isolated from a convenience sample of removed CVAD’s through culture dependent and independent techniques.
-
Health economic evaluation of cost of intervention versus health care utilization and associated costs of diagnosis and treatment of catheter complications
Participant timeline {13}
Participant Timeline is described in Table 1 below.
Participants will be enrolled during the first three months of each four-month period. Each participant
will be followed until they leave the ICU/HAU they reach the end of the 28-day follow-up period, or they die. Data collection for incidence rates of confirmed/suspected CLABSI, catheter occlusion requiring removal, catheter obstruction requiring alteplase use, direct cost related to alteplase use for catheter occlusion, and catheter colonization will continue via the combination of in-person discussion and chart review for the duration of time that the patient has their catheter in place, until they are discharged from the ICU/HAU, or they die, whichever comes first. In the event that a blood culture is taken on either the last day of catheter placement, or the last day that the patient is in the unit, the research coordinator will continue to monitor patient charts for the result of that blood culture, as a bloodstream infection may still be attributed to the catheter.
Data collection in the form of monitoring medical imaging results for incidence rate of catheter-associated venous thrombosis will continue for the entirety of the patients ICU/HAU stay, even if all CVADs have been removed. The Senior Investigators clinical experience indicates that thrombosis from a CVAD in the upper thoracic region is sometimes not discovered until weeks after the CVAD has been removed, as it may take this long for the patient to develop symptoms of thrombosis. As such, the research coordinators will continue to follow medical imaging results of patients who are still in the ICU/HAU but have had all central lines removed and are no longer receiving the study intervention.
Sample size {14}
Results from a one-month observational study conducted at the lead investigator’s site indicate a 20% baseline risk of the composite outcome. The manuscript of this observational study is in production. Analysis included central venous catheters, PICC lines, and hemodialysis catheters. In previous clinical studies, 4% EDTA solution has shown a 40-70% reduction in occlusion (measured by total Alteplase use before vs. after intervention). Assuming a conservative 30% relative risk reduction in the risk of the composite outcome, our sample size calculation is based on an absolute risk reduction from 20% to 14%. A total of 1524 patients need to be recruited to achieve adequate power.
The simulated sample size calculations were performed using R with 10000 simulation runs, which yields a standard error of 0.4% in the estimated power. The calculation assumed a standard deviation of baseline risk across different ICU’s to be 0.02, which corresponded to a within-cluster within-period correlation of 0.0027. The standard deviation between cluster-time periods was assumed to be 0.001. A generalized linear mixed effect model was used to generate and analyze the data. The statistical significance was determined using a t-test with degrees of freedom equal to the number of cluster-periods minus the number of periods minus one (31).
Recruitment {15}
Each participating ICU will have at least 30 beds and be located in large urban centers. Results of an observational study conducted at the Senior Investigator’s 30-bed ICU and step-down unit during December 2020 reveal that 75 patients out of a total of 90 met inclusion criteria to be enrolled in the study. Conservatively, we aim to recruit at least 50 of patients per ICU per month. We do not perceive recruitment challenges as we have obtained waived consent from relevant ethics boards.
With a three-month enrolment window during each period, each site is expected to recruit at least 150 patients per period. As six ICU’s will be participating in this study and each site enrolls for two periods, the enrollment will be 1800 patients in total. Protocol adherence will be deemed a success if there is greater than 95% of patients who are adherent to the entire study protocol.
Assignment of interventions: allocation
Sequence generation {16a}
Randomization involves only the allocation of the six ICUs to the locking solution that will be used during the first period. The randomization will be performed by a statistician at the Centre for Health Evaluation and Outcome Services (CHEOS).
Concealment mechanism {16b}
The statistician performing the randomization will disclose the allocations to only the study Project Manager (PM). The PM will then liaise with each site pharmacy to inform them of their experimental condition so that they can prepare the appropriate locking kits during each period. Blinding of clinicians and research staff will remain throughout the entirety of the trial.
Implementation {16c}
Allocation sequence will be generated by the external statistician. Study participants will be enrolled by designated research coordinators at each site. All patients at each participating site will receive the treatment which is being administered at the site during that period.
Assignment of interventions: Blinding
Who will be blinded {17a}
Trial participants, care providers, outcome assessors, and data analysis will be blinded to the assignment of intervention. The only research personnel who will be unblinded are the statistician performing the randomization, the PM, and the research pharmacists in charge of supplying locking kits used during the study. As both 4% tetrasodium EDTA, and saline and citrate (standard of care) will be placed in opaque syringes, they will not be distinguishable to trial participants, care providers, or outcome assessors to maintain their blinding. Data analysis will occur outside of the unit with the type of intervention received by each patient concealed until the completion of analysis.
Procedure for unblinding if needed {17b}
If unblinding is necessary for patient management (e.g., in the case of an adverse event (AE) for which patient management requires knowledge of treatment assignment), the investigator will be able to obtain this information by getting approval from the study Principal Investigator (PI) to contact a designated individual with access to the allocation list. The study PI and the designated individual will be available 24 hours per day and 7 days per week. Treatment codes will not be broken except in emergency situations that affect clinical care decisions. The investigator requesting unblinding will document and provide an explanation for the request.
Data collection and management
Plans for assessment and collection of outcomes {18a}
All information collected for each patient will be obtained either by reviewing patient charts or engaging in conversations with the treating nurses daily. Members of the research team at each site will extract each participant’s study data from the patient charts and/or resulting nursing conversations and enter the data directly into a secure, electronic database. The lead investigator will instruct each of the study sites regarding data capture procedures on electronic and/or paper CRFs.
Plans to promote participant retention and complete follow-up {18b}
All experimental procedures will be administered by ICU/HAU and research staff during the patients stay. As such, there will be no efforts to promote participant retention.
Patients will be followed by members of the research team until they leave the ICU/HAU, they reach the end of the follow-up period, or they die. After leaving the ICU/HAU, the patients will no longer be part of the study and will not be retained. Patients who are transferred out of the ICU/HAU with a central line in situ will no longer receive the study locking solution and revert to standard of care for the unit to which they have been transferred to. To avoid logistical complications due to the sheer number of participants, patients who leave the ICU/HAU for another hospital unit with their original CVAD still inserted will not be followed. Patients will not be followed past the end of the experimental period.
Data management {19}
Data will be managed by an external data monitoring center. The secure database will be managed according to standard operating procedures and will remain in secure locations on site throughout the entirety of the study.
Confidentiality {27}
In order to maintain patient privacy, data capture tools, study drug accountability records, study reports, and communications will identify the assigned patient number. Confidentiality standards are maintained by coding each patient enrolled in the study through assignment of a unique patient identification number. Patient names or any identifying information will not be included in aggregate data that are transmitted for analysis. Only site enrollment logs maintained by research coordinators in charge of data acquisition will contain any personal identifiers. These logs will remain in a secure location on site for the duration of the study. Only research coordinators will access this information. Additionally, the site Investigators will grant monitor(s) and or its designee access to the patient’s original medical records, including medical history, laboratory studies, and medication administrations, for verification of data gathered and auditing the data collection process. This information will be accessed for the duration of the study for the purpose of data verification and reconciliation.
Plans for collection, laboratory evaluation and storage of biological specimens for genetic or molecular analysis in this trial/future use {33}
A sample of convenience (n=15) of all catheters removed for suspected or confirmed CLABSI or colonization during work hours at the lead investigators site will be collected and used for analysis of biofilm formation and classification of microorganisms. The same number of CVADs removed from patients who no longer require a CVAD with absence of colonization will also be collected to act as a control comparator in this analysis. Upon removal, these CVC will be placed in specimen bags stored at 4℃. These catheters will then be processed by study personnel before shipment to Dr. Aaron White’s lab at the Vaccine and Infectious Disease Organization-International Vaccine Centre (VIDO-InterVac) for further analysis. These will include culture-dependent and culture-independent classification of microorganism colonies and analysis of biofilm, according to internal Standard Operating Procedures.
Statistical methods
Statistical methods for primary and secondary outcomes {20a}
Data will be analyzed following the intent-to-treat principle. The primary outcome will be analyzed using a zero-inflated negative binomial (ZINB) model that accounts for clustering within ICU (32). The ZINB model takes into account: (1) the large percentage (~80% expected) who will not have any event and (2) the large variability in the number of events among participants who experience at least one event (individuals who experience one event are more likely to experience additional events). The (log-)time-at-risk will be entered into the model as an offset term. The average treatment effect will be obtained by regression standardization of the composite incidence rates predicted by the ZINB model (32). Standard errors will be obtained using bootstrapping.
The model will adjust for the study period (first or second) and may adjust for the following individual-level baseline characteristics: age, sex at birth, BMI, APACHE IV score, length of pre-ICU hospital stay (days), presence of multiple CVADs per patient, any suppression of the immune system, defined as: post-organ transplantation, Acquired Immunodeficiency Syndrome [AIDS], neutropenia [<1000 absolute neutrophils], corticosteroids [>20 mgs/day of prednisone or equivalent for more than 6 months], or currently taking immunosuppressive agents (i.e. Imuran, etc) at time of enrollment into study, site of CVAD insertion (femoral, jugular, subclavian, right or left side), type of catheter used, any catheter additives, as they may confound the result (such as antibiotic impregnation), having a pre-existing central line upon admission to the ICU, and this line had already received a lock., concurrent systemic antibiotic use unrelated to central line infection, concurrent systemic anti-fungal agent use unrelated to central line infection, concurrent systemic administration of anti-coagulants unrelated to catheter occlusion, type of hemodialysis (if applicable), previously experienced CLABSI or CVAD occlusion.
The secondary outcomes will be analyzed similarly to the primary analysis using generalized mixed effects models (to account for clustering within the unit) appropriate to the type of outcome. Further analysis of results will include calculation of cost attributed to healthcare utilization costs between groups. A complete statistical analysis plan will be developed prior to unblinding and analysis of the data.
Interim analyses {21b}
Due to the cluster-crossover design, there will not be sufficient data from both treatment conditions to enable within-center comparisons until near the end of the study. As such, interim analysis for early efficacy or futility would not be informative. Preliminary data will be analyzed only for safety (by tabulating the total percentage of AEs observed) and recruitment rates (by reviewing the total number of patients enrolled and comparing this to predicted rates).
Methods for additional analyses (e.g. subgroup analyses) {20b}
Subgroup analysis will be conducted for each participating ICU to assess potential heterogeneity of treatment effects across sites using a generalized linear model stratified by ICU. Additionally, due to standard of care locking differing between CVCs, IVADs, and PICCs (locked with saline) as compared to hemodialysis catheters (locked with citrate), the authors will perform analysis of the hemodialysis population both together and separately from the rest of the patient population.
During the COVID-19 pandemic, the expansion and shuffling of ICU and HAU patients to different units of the hospital poses a logistical problem in running of experimental protocol (as this would involve having to provide education and oversight of nursing staff beyond the ICU/HAUs). As such, patients who are still designated as ICU/HAU care level, but physically located in another unit, will not receive the locking intervention, and will be followed by Research Coordinators for their location in the study. Should they be repatriated back to the physical spaces of the ICU/HAU, they will resume receiving the study locking protocol. Because these patients would have experienced an interruption to experimental protocol, they will also be analyzed in a separate subgroup.
Finally, recent studies have also provided evidence of higher risk of catheter occlusion and catheter-associated venous thrombosis among COVID-19 patients, due to the hyper-inflammation and hyper-coagulation associated with the virus (33-35). As such, we will be collecting data about COVID-19 status of each enrolled patient. This subgroup will also be analyzed separately, to determine if there are any differences among the COVID-19 population.
Methods in analysis to handle protocol non-adherence and any statistical methods to handle missing data {20c}
We do not expect any issues related to protocol non-adherence as detailed education about trial procedures will be given to all clinical staff involved prior to study start. No cross-over is expected only one of the two locking solutions will be available at each site during each period.
We do not expect to have a significant number of missing data points as we are not including any measures that need to be collected outside of standard of care. Multiple imputation will be used to fill in missing data on covariates, if needed. Should any outliers arise, we will conduct sensitivity analysis to assess their impact on the conclusions.
Plans to give access to the full protocol, participant level-data and statistical code {31c}
Participants may request access to de-identified data. Patient-specific data will be available at the completion of data analysis.
Oversight and monitoring
Composition of the coordinating centre and trial steering committee {5d}
The daily operations of this project will be managed by site Research Coordinators. Each center will appoint a Research Coordinator responsible for enrollment and daily data collection, as well as recording of AEs. Any AEs and enrollment numbers will be communicated to a separate trial steering committee on a bi-weekly basis by each site Research Coordinator. The trial steering committee will consist of co-investigators and will meet via teleconference on a bi-weekly basis, starting at 4 weeks prior to study start to discuss trial progress and any necessary protocol amendments. The last meeting will be held 4 weeks after study end. The trial steering committee will also be responsible for producing the final paper for publication. The data collection database will be according to internal Standard Operating Procedures.
Composition of the data monitoring committee, its role and reporting structure {21a}
4% EDTA solution has a very low possibility of harm as it is already Health Canada approved, is marketed and sold, and patients have minimal systemic exposure as it will be aspirated prior to the lumen being used. As such, investigators feel a standing DMC is not necessary. If at any time, more than 3 serious adverse events are recorded, a DMC will be assembled to review trial safety.
Adverse event reporting and harms {22}
Information about any AEs will be collected with routine data collection. AEs will include (1) local site irritation or redness following accidental administration of the catheter lock into the patient circulation, (2) Dysgeusia following accidental administration of the catheter lock into the patient circulation, (3) Paresthesia following accidental administration of the catheter lock into the patient circulation, (4) pain upon accidental administration of the catheter lock into the patient circulation, (5) ‘head rush’ following administration, (6) blood rush to face and lips following administration, (7) hypocalcemia due accidental administration of the catheter lock into the patient circulation, and (8) any other adverse reaction observed during accidental administration of the catheter lock into the patient circulation.
Frequency and plans for auditing trial conduct {23}
Monitoring and auditing procedures developed by the lead site investigator, or its designee will be followed, in order to comply with Good Clinical Practice and International Council for Harmonization of Technical Requirements for Pharmaceuticals for Human Use (ICH) guidelines. Inspection of the database for consistency completeness and clarity, in comparison to with source documents will be performed. Any clarification of administrative matters will also be completed in this process. The study will be monitored by the lead site or its designee. Monitoring will be done by personal or virtual visits from a representative of the lead site.
All AEs will be reviewed by the principal investigator. All AEs will be reported to the relevant ethics boards within 72 hours of their occurrence. The relevant ethics boards will perform audits as per standard operating procedures. The independent physician and members of the relevant ethics boards are both independent from the investigators and the sponsor.
Plans for communicating important protocol amendments to relevant parties (e.g. trial participants, ethical committees) {25}
Research Coordinators and Investigators at each site will meet on a bi-weekly basis to discuss ongoing trial progress. Any important protocol modifications will be discussed and communicated during these meetings. Study materials will also be updated with any new information and given to patients and caretakers as necessary. Any changes to the protocol must be approved by the relevant ethics boards prior to implementation.
To keep the public informed on clinical trials in a timely manner, and to comply with applicable laws, regulations and guidance this study is registered with clinicaltrials.gov. This entry will be maintained by members of the research team to always display up to date information about the trial.
Dissemination plans {31a}
Completion and results from this study will be used to fulfill thesis requirements for one graduate-level student. Plans for publication include submission of the thesis, publication in an academic journal of the student/supervisors choosing, as well as oral dissemination of results in the form of a formal thesis defense as well as submission to conference proceedings. Conference presentations will serve to disseminate results to academic and clinical professionals.
In addition, patient partners will be engaged to identify target audiences outside of the medical research community, and direct the creation of meaningful knowledge dissemination materials for these audiences, such as infographics, brochures, or digital media.
If successful, results of this trial can be used to inform key decision makers of available improvements to CVAD practices and ICU budgets through use of 4% tetrasodium EDTA. These improvements may lead to changes to standard of care to include novel CVAD lock solution technology.