key: cord-0877804-yyghy0gb
authors: Maurer, Lydia R.; Luckhurst, Casey M.; Hamidi, Arzo; Newman, Kelly A.; Barra, Megan E.; El Hechi, Majed; Mokhtari, Ava; Breen, Kerry; Lux, Laura; Prout, Laura; Lee, Jarone; Bittner, Edward A.; Chang, David; Kaafarani, Haytham M.A.; Rosovsky, Rachel P.; Roberts, Russel J.
title: A low dose heparinized saline protocol is associated with improved duration of arterial line patency in critically ill COVID-19 patients
date: 2020-09-02
journal: J Crit Care
DOI: 10.1016/j.jcrc.2020.08.025
sha: 72eb0c7b0a302501b627d37eac2abdd28a2bd95f
doc_id: 877804
cord_uid: yyghy0gb

PURPOSE: Critically ill patients with Coronavirus Disease 2019 (COVID-19) have high rates of line thrombosis. Our objective was to examine the safety and efficacy of a low dose heparinized saline (LDHS) arterial line (a-line) patency protocol in this population. MATERIALS AND METHODS: In this observational cohort study, patients ≥18 years with COVID-19 admitted to an ICU at one institution from March 20–May 25, 2020 were divided into two cohorts. Pre-LDHS patients had an episode of a-line thrombosis between March 20–April 19. Post-LDHS patients had an episode of a-line thrombosis between April 20–May 25 and received an LDHS solution (10 units/h) through their a-line pressure bag. RESULTS: Forty-one patients (pre-LDHS) and 30 patients (post-LDHS) were identified. Baseline characteristics were similar between groups, including age (61 versus 54 years; p = 0.24), median Sequential Organ Failure Assessment score (6 versus 7; p = 0.67) and systemic anticoagulation (47% versus 32%; p = 0.32). Median duration of a-line patency was significantly longer in post-LDHS versus pre-LDHS patients (8.5 versus 2.9 days; p < 0.001). The incidence of bleeding complications was similar between cohorts (13% vs. 10%; p = 0.71). CONCLUSIONS: A LDHS protocol was associated with a clinically significant improvement in a-line patency duration in COVID-19 patients, without increased bleeding risk.

antithrombin deficiency, pregnancy, obesity, diabetes, oral contraceptive use, malignancy, and peripheral vascular disease), Sequential Organ Failure Assessment (SOFA) score on the day of observation period start, and history of recent hospitalization or surgery in 30 days prior to ICU admission. Systemic administration of anticoagulation (sometimes referred to as "therapeutic anticoagulation") at any point during the observation period was recorded, along with the specific agent that was used, the indication for anticoagulation, and the target anti-Xa or PTT level associated.

Baseline coagulation laboratory values were also recorded [international normalized ratio (INR), partial thromboplastin time (PTT), fibrinogen, anti-Xa level, platelet count, C-reactive protein (CRP), D-Dimer, and ferritin level]. If lab data were missing, this was excluded from the analysis.

Data collection began for each patient with the index a-line. The "index a-line" was the a-line in place at protocol start in post-LDHS patents, and the a-line placed following the first a-line thrombosis event in pre-LDHS patients. The observation time was defined as the duration of LDHS administration in post-LDHS patients, and the total duration from index a-line placement to final a-line removal in pre-LDHS patients. For each group, the total number of a-lines placed prior to the index a-line and the location of the index a-line were documented, along with all a-lines replaced or re-wired during the observation period.

The primary outcome was duration of index a-line patency, measured from initiation of the LDHS protocol (post-LDHS patients) or placement of index a-line (pre-LDHS patients) until an endpoint of documented a-line thrombosis requiring a-line re-wiring or replacement. If there were no thrombotic episodes requiring a-line re-wiring or replacement, the duration of time was censored at LDHS protocol discontinuation, J o u r n a l P r e -p r o o f Journal Pre-proof permanent a-line removal, or patient death. In order to identify episodes of a-line thrombosis, chart review was performed looking for re-wiring or replacement, preceded by documentation of one of the following: 1) Severe arterial waveform dampening that did not improved with flushing or repositioning, 2) Inability to withdraw blood from the catheter, and/or 3) Catheter malfunction requiring replacement with documented thrombosis on bedside ultrasound of the vessel. If an a-line was replaced, but a reason could not be identified, this was included as a positive event. A-line removal for other reasons (i.e. infection, accidental dislodgement, kinking, etc.), were excluded from the primary outcome analysis. The secondary outcome was incidence of bleeding complications, divided into major bleeding, clinically relevant non-major bleeding (CRNMB), and minor bleeding, as defined by the International Society for Thrombosis and Hemostasis [13] . Development of heparin-induced thrombocytopenia (HIT) with a positive heparin-PF4 antibody test was also recorded in the post-LDHS patients.

Descriptive statistics were performed and reported as median and interquartile range (IQR), or numbers and percentages as appropriate.

Categorical data were evaluated using the Chi-square or Fisher's exact test, as appropriate, and continuous variables were evaluated with Wilcoxon rank sum test. The median duration of index a-line patency was initially calculated in each group using a combined endpoint including the first of: a-line thrombosis, permanent a-line removal, or death. Kaplan-Meier methodology was used to compare differences in a-line patency duration between the groups. In this analysis, failure was defined as a-line thrombosis, and all patients were censored at the 20-day time point.

Additional censoring events included permanent removal of the a-line for reasons other than thrombosis, or death while the a-line was still patent.

A log rank test was performed to compare the Kaplan Meier curves of the two groups. Cox proportional hazard regression analysis was conducted to define independent predictors of arterial line thrombosis, adjusting for the following variables: age (categorized as 18-39 years, 40-64 years, and J o u r n a l P r e -p r o o f Journal Pre-proof 65+ years), sex, obesity (BMI<30kg/m 2 vs. BMI≥30 kg/m 2 ), history of thrombosis, presence of the most common thrombotic risk factors (diabetes, malignancy, and smoking status), recent surgery (within 30 days), SOFA score (as a continuous variable), index arterial line location, total arterial lines prior to index arterial line, systemic anticoagulation use, and LDHS protocol use. A two-sided p-value of ≤0.05 was used to denote statistical significance. All statistical analyses were performed using STATA® release 14.2 (StataCorp LLC, College Station, TX, USA). GraphPad Prism 8 (GraphPad Software, LLC, San Diego, CA) was used for Kaplan Meier figures.

A total of 71 patients were included, with 30 patients in the post-LDHS cohort, and 41 patients in the pre-LDHS cohort. Baseline characteristics of both cohorts are described in Table 1 . Compared to pre-LDHS patients, post-LDHS patients were similar in age (median 61 vs. the pre-LDHS group who specifically reported a history of DVT (N=2), PE (N=2), and ischemic stroke (N=1) (p=0.07). Of those patients who received prophylactic anticoagulation during the study period, there was no significant difference in the agent chosen (subcutaneous enoxaparin vs. subcutaneous heparin) between the pre-and post-LDHS groups (38% heparin versus 55% heparin; p=0.28). Table 2 missing; pre-LDHS 7 missing). All patients were followed to the first event of the following: death, ICU discharge, or the 20-day time point.

Median duration of a-line patency, bleeding complications, and deaths are indicated in Table 3 

Given the risk of confounding by the five patients in the pre-LDHS group with a past history of thrombosis, a sensitivity analysis was performed with these five patients excluded. In this analysis, the significant effect was maintained, with a median (IQR) duration of a-line patency of 8.5 days On Cox proportional hazard regression analysis, independent predictors of a-line thrombosis were obesity and not being on the LDHS protocol (Table 4 ). Obese patients (BMI≥30 kg/m 2 ) had more than twice the risk of a-line thrombosis than patients with BMI<30 kg/m 2 (hazard ratio 2.28; 95% CI 1.07-4.84; p=0.03). Post-LDHS patients had an 87% thrombosis risk reduction compared to pre-LDHS patients (hazard ratio 0.13, 95% CI 0.05-0.34; p<0.001). Systemic anticoagulation did not significantly decrease the risk of index a-line thrombosis (hazard ratio 0.61; 95% CI 0.27-1.38; p=0.23).

The incidence and severity of bleeding complications, stratified by systemic anticoagulation and post-LDHS versus pre-LDHS group, are available in Table 3 . While there were no major bleeding complications in the post-LDHS group, three CRNMB complications (all of which occurred in patients receiving systemic anticoagulation) did occur. Among pre-LDHS patients, two patients had major bleeding complications. Table 5 .

There were no significant differences in coagulation labs before versus after LDHS protocol initiation in post-LDHS patients, including INR, PTT, or platelet count ( Table 6) . None of the patients in the post-LDHS group developed HIT.

In critically ill patients with COVID-19, a LDHS protocol was associated with increased duration of a-line patency in patients with a history of a-line thrombosis. Over 90% of post-LDHS a-lines were still patent at day 5, compared to less than 40% of catheters in the pre-LDHS group. There were no significant differences in coagulation labs after LDHS initiation, nor were there increased bleeding complications in the post-LDHS cohort.

The issue of arterial access is particularly important for critically ill patients admitted with COVID-19. Across multiple studies, median ICU length of stay for COVID-19 patients is over two weeks, with patients spending a median of 10-16 days on the ventilator, and 28-47% of patients requiring prone positioning [14] [15] [16] . The predominant mechanism of a-line thrombosis is intravascular thrombosis immediately proximal to the catheter tip [17] . Local mechanisms of endothelial damage have been proposed to explain catheter-associated thrombosis, including shear stress from guidewire advancement and catheter flushing, fibrin accumulation due to local turbulent flow, and local trauma on the vessel wall by the catheter [18] [19] [20] [21] [22] . In Prior to the COVID-19 outbreak, a number of studies have examined the use of LDHS protocols in ICU patients to improve a-line patency, with mixed results [6] [7] [8] [9] [10] [11] [12] 25, 26] . The proposed mechanism of the LDHS protocol is infusing the anticoagulant directly to the site of potential clot formation proximal to the catheter tip. In a randomized-controlled trial of 300 pediatric ICU patients, patients who received sodium chloride 0.9% had more than three times the risk of a-line thrombosis versus patients who received the LDHS protocol at 2-5 units/hour [7] . In a multicenter, randomized-controlled trial of over 5,000 adult ICU patients, a LDHS protocol was associated with increased a-line patency, along with longer catheters (>2 inches), systemic anticoagulation, femoral catheter placement, and male sex [26] . Conversely, in adult cardiac surgery patients, a LDHS infusion was not associated with an increased rate of patency [12] . A Cochrane review committee in 2014 deemed the evidence J o u r n a l P r e -p r o o f Journal Pre-proof limited and biased, and were unable to perform a meta-analysis [25] . This body of evidence has led to heterogeneous protocol use, often varying within hospitals by dose and patient population. While doses varied between studies, the Cochrane authors did conclude that while lower doses had varied effectiveness, one study did show significantly increased patency duration with a dose of 4 units/mL (run continuously for a total of 12 units/hr) [6] . We believe that using a dose of 10 units/hr and targeting therapy to patients with COVID-19 who had already thrombosed at least one a-line, our selected population was likely at higher risk of a-line thrombosis, and more likely to benefit from treatment.

There are several limitations of this study. First, this was an observational study that depended on chart review. We used proxies of thrombosis including a-line dampening and inability to draw back, as ultrasound evidence was inconsistently documented. Severe a-line dampening and inability to withdraw blood have both been described as appropriate proxies of thrombosis in previous randomized-controlled trials of a-line patency, even in the absence of ultrasound confirmation [8] . In one study, 84% of patients with either an overdampened wave form, sluggish backflow, inability to withdraw blood or inability to flush were found on bedside ultrasound to have intravascular thrombus adjacent to the catheter tip [17] . Furthermore, arterial catheter type and placement technique (e.g. "through-and-through" insertion technique, micropuncture, etc.) were not included, due to inconsistent documentation. That said, in a study comparing the risk of thrombosis in radial arteries cannulated using a "through-and-through" methodology (where the access needle is advanced through the back wall of the artery) compared to those cannulated directly into the artery showed no difference in patency [27] . Finally, a-line re-wires were also not always documented and these events may not have been fully captured despite review of nursing, procedure, and provider notes to confirm as many procedures as possible.

To the authors' knowledge, this is the first study to evaluate the implementation of a LDHS protocol in patients with COVID-19 who developed a-line thrombosis. The protocol was implemented across previously established ICUs along with "surge" ICUs (developed in response to the surge of critically ill patients with COVID-19), highlighting its streamlined approach and ease of use. Furthermore, we did not observe any J o u r n a l P r e -p r o o f major bleeding complications in the LDHS cohort. While a study of cardiac surgery patients found a small but statistically significant increase in activated clotting time (ACT) and PTT with LDHS [12] , in our study, there were no differences in coagulation markers (ACT was not followed).

As hospitals continue to care for critically ill patients with COVID-19 now and in the future, there are multiple benefits to a low-risk intervention that prolongs a-line patency and limits the need for invasive procedures. For institutions faced with equipment and medication shortages during high-volume admission periods, implementation of the LDHS protocol in patients with COVID-19 who develop a-line thrombosis has the potential to maintain a-line patency while reducing the need for systemic anticoagulation. The results of this study suggest that a LDHS protocol is a safe and effective therapeutic option to implement more broadly in patients with COVID-19 who develop a-line thrombosis.

A LDHS protocol is associated with an increased duration of a-line patency in critically ill patients with COVID-19, without an increased risk of bleeding complications. Given the protocol's success in patients who have already demonstrated thrombosis, further prospective investigation is required to determine if this protocol could benefit a broader population of critically ill patients with COVID-19, before development of initial a-line thrombosis.

We appreciate the multidisciplinary assistance from the pharmacy, ICU, nursing, hematology, surgery, and anesthesia departments, who all supported the implementation of this quality improvement effort and its evaluation.

This research did not receive any specific grant from funding agencies in the public, commercial, or not-for-profit sectors. 

Gastrointestinal Complications in Critically Ill Patients With COVID-19

Venous and arterial thromboembolic complications in COVID-19 patients admitted to an academic hospital in

Incidence of thrombotic complications in critically ill ICU patients with COVID-19

Design and Impact of a COVID-19 Multidisciplinary Bundled Procedure Team

Arterial and vein related catheters thrombosis in a patient with COVID-19 : a case J o u r n a l P r e -p r o o f report

Comparison of normal saline and heparin solutions for maintenance of arterial catheter patency

The efficacy of heparinization in prolonging patency of arterial and central venous catheters in children: A randomized double-blind trial

Finding a solution: Heparinised saline versus normal saline in the maintenance of invasive arterial lines in intensive care

Heparinized and saline solutions in the maintenance of arterial and central venous catheters after cardiac surgery

Heparinized solution vs. saline solution in the maintenance of arterial catheters: A double blind randomized clinical trial

Heparinized saline versus normal saline in maintaining patency of the radial artery catheter

A comparison of heparinised and non-heparinised normal saline solutions for maintaining the patency of arterial pressure measurement cannulae after heart surgery

Definition of clinically relevant non-major bleeding in studies of anticoagulants in atrial J o u r n a l P r e -p r o o f fibrillation and venous thromboembolic disease in non-surgical patients: Communication from the SSC of the ISTH

Respiratory Pathophysiology of Mechanically Ventilated Patients with COVID-19: A Cohort Study

Covid-19 in Critically Ill Patients in the Seattle Region -Case Series

Clinical course and outcomes of critically ill patients with SARS-CoV-2 pneumonia in Wuhan, China: a single-centered, retrospective, observational study

Impact of intravascular thrombosis on failure of radial arterial catheters in critically ill patients: a nested case-control study

Histopathologic changes of the radial artery wall secondary to transradial catheterization

Long-term radial artery cannulation: effects on subsequent vessel function

Patency of radial arterial catheters

The effects of arterial flow on platelet activation

PTT (seconds)