Management of catheter-related right atrial thrombus in hemodialysis: a systematic review

Abstract

Objective

Hemodialysis catheter-related right atrial thrombus (CRAT) is a rare and fatal complication related with catheter. Treatment recommendations are controversial. We reported our institution’s recent three cases in managing CRAT and review and analyze the reported cases of CRAT in hemodialysis patients.

Methods

A systematic search of the PubMed, Embase and Web of Science databases was conducted to identify the therapy and outcome data in hemodialysis CRAT patients.

Results

From 1975 to November 2023, a total of previous 144 cases which reported in the literatures and three new cases in our institution of CRAT in hemodialysis were included and analyzed. Overall mortality was 18.1% (26/144). Most of cases can be detected by echocardiography. Thirty-three patients had no treatment, except for catheter removal, replacement or antibiotics, but eleven of them have died. Thrombolytic therapy was adopted in 14 cases but only nine cases was successful, the remaining cases need to further therapy. Eventually, 71 cases have been treated by anticoagulation and 34 cases received thrombectomy.

Conclusions

We recommend that the replacement of the catheter and anticoagulation combined with thrombolysis is a preferred therapy. Thrombectomy should be considered when other methods fail or new complication happened. Thrombolysis alone has a low success rate but may be useful in combination with anticoagulant therapy.

Introduction

Hemodialysis catheters are commonly used among patients with end-stage renal disease on hemodialysis patients. Catheter-related right atrial thrombus (CRAT) is a rare but fatal complication of hemodialysis catheter in children and adults. The incidence rate of CRAT in reported studies varies from 2 to 29% [1, 2]. This is are underreported phenomena due to the largely asymptomatic patients with CRAT, the relative insensitivity of current methods for discovering CRAT, and the spontaneous dissolution of some thrombi [3–5]. CRAT can lead to serious consequences such as pulmonary embolism, septic emboli, cardiac dysfunction or even systemic embolization through a patent foramen ovale. It has a high mortality rate of up to 45% [6, 7]. With the more wide-spread usage of hemodialysis catheters over the years, an increased incidence of CRAT can be expected, but a consensus has not been reached about the optimal treatment and management for hemodialysis patients with CRAT [3, 8, 9].

Stavroulopoulos et al. recommend catheter removal should be the first choice in the management of a CRAT [3]. Whereas, this vascular line is vital for the hemodialysis patients alternative vascular accesses, rescue of central venous catheter should be taken. As a result, Yang H et al. recommended that maintenance of hemodialysis (HD) by replacing catheters and providing oral anticoagulation/antiplatelet therapies may be an effective strategy for treating hemodialysis patients with CRAT [10]. But it may lead to recirculation rate of 17% with good dialysis adequacy. Therefore, the aim of this study is to present three new cases with CRAT who were treated through replacing catheters without a higher recirculation rate in our center and analysis the reported cases in the literatures for comparing diagnosis, treatment options and outcome.

Method

Case selection

The three CRAT cases were collected from our vascular access centers from 2019 to 2023 who were dialyzed by a double-lumen, tunneled dialysis catheter. Clinical data and radiological images were obtained (Table 1). All patients were hospitalized because of catheter dysfunction with no other symptom. The right atrial thrombus was detected by routine echocardiography (Fig. 1) and all the cases have undergone the computerized tomography examinations. All patients were maintained on regular HD without thrombolysis, and were followed up by routine echocardiography once every three months. Two of the patients have survived in the follow-up time of 6 ~ 12 months with oral anticoagulation (Rivaroxaban 10 mg qd), remaining one case died of gastrointestinal hemorrhage due to cervical cancer metastasis (Patient No.1 of Table 1).

Table 1.

Characteristics of the cases in our vascular access center

Patient No.	Age(yr)/ gender	Cause of ESRD	Chief complaint	Blood culture	Duration of Catheter Use(years)	Location of Catheter Tip	Imaging method	Patent foramen ovale	Maximal Diameter of RAT(cm)	Treatment
1	56/female	hypertension	catheter dysfunction	None	4	RA	echocardiography	No	5.4	Catheter replacement + oral anticoagulation
2	86/female	hypertension	catheter dysfunction	None	1	RA	echocardiography	No	3.6	Catheter replacement + oral anticoagulation
3	65/female	unknow	catheter dysfunction	None	1	SVC	echocardiography	No	2.8	Catheter replacement + oral anticoagulation

Fig. 1.

Red arrows show the size of catheter-related right atrial thrombosis (CRAT) and the relationship between the thrombosis and catheter: (a) Transthoracic echocardiography shows a thrombus 2.8 cm×2.6 cm (arrow) in the right atrium. (b) DSA reveals the thrombosis is attach to the catheter

Operation procedure

Based on the diagnosis of CRAT, catheters were replaced over a guidewire in situ guided by digital subtraction angiography without disturbing the thrombosis. The operator maneuvered the guide wire to directly reach the inferior vena cava, pulled out the old catheter, and successfully inserted a new catheter (Palindrome, Covidien Health Costa Rica, USA). The entire procedure had no effect on thrombosis. Upon re-examination by echocardiography, no change was found in the right atrial thrombosis. Due to the absence of any interference from atrial thrombosis, the blood flow of the new catheter dialysis reached the standard rate, with a kt/v value of 1.1 ~ 1.2.

Systematic review

Search strategy and study selection

A systematic search in PubMed databases was performed for publications from 1975 to November 2023, utilizing the terms “catheter”, “hemodialysis”, “right atrium”, “thrombus” and “thrombosis”. Studies with CRAT in hemodialysis patients about management and outcome were considered eligible. Studies about right atrial thrombus not associated with a hemodialysis catheter, deep venous thrombosis and tumor thrombus were excluded. Forty-eight studies, 40 case reports [3, 4, 8, 9, 11–44] and 8case series (≥ 5 cases) [1, 10, 45–50], were included in our study. Data for each patient included age, sex, cause of end-stage renal disease, duration of catheter use, location of the hemodialysis catheter’s tip, presenting symptoms, maximal diameter of the thrombus, imaging method for the diagnosis of thrombus, presence of foramen ovale(PFO), catheter’s management after CRAT diagnosis, treatment strategies for thrombus (no treatment, thrombolysis, anticoagulation, surgical thrombectomy), prognosis (survival or death). Not all of the studies included all these data; however, they were included in our study because they provided the data of treatment and prognosis. Furthermore, only one patient who was adopted percutaneous thrombectomy in these studies was classified as a surgery [18].

Statistical analysis

Subject characteristics are presented as mean ± SD. Analysis was performed using SPSS version 25 (SPSS Inc., Chicago, Illinois). Due to the small sample size and the not normal distribution of some of the observations, the non-parametric Mann–Whitney–Wilcoxon rank-sum test was used for comparison of continuous variables between independent groups. Chi-square test was used for comparison of binary variables. P value less than 0.05 was considered statistically significant.

Result

Presentation and mortality

A total 144 cases of CRAT in hemodialysis (not including our patients) were included in this study. The data about the patients was showed in Table 2. The most common causes of end-stage renal disease were Diabetes mellitus and the patients who have Diabetes mellitus more easily suffered from infective catheter-related right atrial thrombi (P = 0.01). The catheter was located in the subclavian vein (30 patients), internal jugular vein (51 patients) and femoral vein (2 patient), unknow (62 patients). The tip of catheter was located in the right atrium (68 patients) or in the right atrium/superior (or inferior) vena cava junction (26 patients), unknow (51 patients). Thrombus can be detected with echocardiography for the most cases. In five patients, diagnosis was made by combined with magnetic resonance imaging [4, 13, 15, 26, 43] and in twenty-three with computerized tomography [10, 30, 31, 34, 37, 41]. 63 patients had no presenting symptoms in the non-infective CRAT patients and they were diagnosed during assessment of a dysfunctional catheter or in routine echocardiography screening. In infective CRAT patients, 58 patients have a symptom such as fever, chest pain or chill etc. Complications contained pulmonary emboli (15 patients), tricuspid regurgitation or partial obstruction of the valve from the thrombosis (16 patients), right heart failure (2 patients), endocarditis (4 patients), cardiogenic or septic shock (11 patients).

Table 2.

Characteristics of the reported cases^a

	Non-infective catheter-related right atrial thrombi n = 84	infective catheter-related right atrial thrombi n = 59	P
Age(years)	49.29 ± 20.78	49.17 ± 18.49	0.972
Male/female	24/60	26/33	0.06
Cause of ESRD
Diabetes mellitus	19	25	0.01
Hypertension	7	4	0.98
Glomerulonephritis	8	5	0.83
IgA nephropathy	3	1	0.877
Polycystic kidney disease	7	1	0.183
Others	16	11	-
Unknow	24	12	-
Duration of Catheter Use (months)	17.63 ± 49.81	33.02 ± 86.95	<0.001
Location of Catheter Tip
RA	46	22	-
SVC/RA junction	15	11	-
unknow	23	26	-
Maximal Diameter of RAT(cm)	2.42 ± 1.35	2.85 ± 1.49	0.103
Imaging method
Echocardiography	79	30	-
MRI and echocardiography	2	3	-
CT and echocardiography	3	20	-
Patent foramen ovale	2	1	-
Bacteraemia	0	58	-

^a n refers to the number of patients in each group. However, as not all of the studies had data on all of the variables studied, n is provided for every individual variable in each group. RA: right atrium, SVC: superior vena cava

Overall mortality was 18.1% (26/144, including the 8 cases of non-infective CRAT patients and 18 cases of infective CRAT patients). Duration of catheter use in the patients with infective CRAT was longer than that of the patients with non-infective CRAT (P<0.001). Diabetes mellitus in the group of infective CRAT patients were more than that of non-infective CRAT patients (P = 0.01).

Management methods

We divided the treatments into four categories. Some patients received more than one treatment as we described in the following paragraphs with more details.

No Anticoagulant/Thrombolysis treatment group

Thirty-three patients received no treatment, except for catheter removal or replacement and antibiotics for bacteraemia, eleven of them have died [11, 14, 19, 23, 46, 49], including one case had refused treatment [11].

Systemic thrombolysis group

Systemic thrombolysis was performed in 14 patients (13 as initial treatment) [1, 20, 25, 29, 33, 37, 45, 48], six of them received recombinant tissue plasminogen activator (rtPA) and another five urokinase, remaining cases have no said about the kind of thrombolytic agent. Seven of them combined it with anticoagulation treatment. Nine patients were successful. One of them had pulmonary emboli and a large thrombus which extended from the right atrium to the right ventricle through a patent foramen ovale [20]. All the treatment with combination of anticoagulation and thrombolysis were successful with dissolution of the thrombus. The remaining patients needed further treatment.

Anticoagulation treatment group

Anticoagulation was the preferred treatment in 63 patients [1, 3, 4, 8, 10, 15, 17, 22, 23, 26, 28, 29, 32, 34, 38, 39, 41, 44, 45, 47–50]; however, it was also the only option for many of them, as they were not suitable for surgical thrombectomy. In fifteen patients, treatment failed; six of these fifteen patients survived included one after obtaining systemic thrombolysis [1], one with no further treatment and the four other after successful surgical thrombectomy [29, 39, 41, 49]. The remaining nine patient died [4, 22, 23, 38, 49, 50]. For the most patients who were successful with anticoagulation, it was performed from 20 days to 10 months with a target INR of 2–3 until complete dissolution of the thrombus [1, 3, 8, 15, 17, 22, 28, 32, 45, 48, 50]. Prolonged (50 months) and enhanced anticoagulation (INR 3–4) was not successful for the case of persistent CRAT [4].

Thrombectomy group

Thrombectomy means surgical thrombectomy on cardiopulmonary bypass which need to open heart surgery. Surgical thrombectomy of the thrombosis was preferred method in 26 patients and only three cases died after operation [22, 49]. For another patient, percutaneous intravascular removal of the thrombosis was successfully performed with a large (6 × 2 cm) CRAT [18]. Thrombectomy was also carried out in eight patients after failure of medical therapy [4, 22, 25, 29, 38, 39, 41, 45]. Three of them died postoperatively [22, 38, 45]. In nine patients, anticoagulation was performed after the surgical removal of the thrombosis [4, 15, 23–25, 32, 35, 41, 42].

Catheter management

For a total of 75 patients, the dialysis catheter was removed at diagnosis, at surgery, or after therapeutic anticoagulation was achieved and 63 cases of them survived. In 34 patients including our three cases, the catheter was exchanged, all survived with oral anticoagulation treatment, including nine cases with bacteraemia [1, 10]. In 38 patients, the catheter was left in place and not removed for a period of time, 28 survived. In these cases, four of them had bacteraemia [45, 49], two died afterwards.

Compared with different management approach according to presence of complications and survivors (Table 3)

Table 3.

Characteristics of reported cases according to final treatment received

	No treatment	Thrombolysis	Anticoagulation	Surgery	P
Survivor	22	14	54	28	0.305
Non-survivors	11	0	9	6
Complication	7	5	22	14	0.317
Emboli	0	3	9	3
Endocarditis	2	0	1	1
Other cardiac	2	0	9	7
Shock	3	2	3	3

Complications were showed in 22 patients in the anticoagulation group, in 14 patients in the thrombectomy group, in 5 patients in the thrombolysis group and in 7 patients in the no treatment group. The difference was not statistically significant (P = 0.317). In 15 patients, CRAT was accompanied by pulmonary embolism. Two of them were treated by systemic thrombolysis and survived finally included one as initial treatment [20] and another after failure of anticoagulation [1]. Eight cases were treated initially with anticoagulation. In two of them, treatment was successful [23, 28]. For the another two, one died thirteen months later [4] and the other one survived after thrombectomy [4]. The remaining four patients remained asymptomatic with organized thrombi during the treatment of anticoagulation. In four patients, pulmonary embolism was detected in the course of anticoagulation: one patient survived successfully after systemic thrombolysis [1], remaining three cases died included one case after thrombectomy [4, 22]. Finally, one case received initially thrombectomy successfully [21]. Four patients were accompanied by endocarditis. Two cases died with no treatment [11, 14] and the other two survived included one after successful anticoagulation [23] and the other after thrombectomy [16]. Compared with the survivors in four treatment methods, no statistical differences were detected (P = 0.305).

Discussion

Our study addressed the diagnosis, presentation, therapy and prognosis of CRAT in hemodialysis patients and extends the present knowledge on the occurrence of this syndrome in hemodialysis patients. We detected that CRAT is a rare complication through analyzing 144 cases of CRAT in hemodialysis patients, with the wide use of central dialysis catheters in hemodialysis patients. Incidence of CRAT is about 5.4% as reported by a retrospective study for hemodialysis patients [1]. However, this may be an underestimated result by the reason of potential asymptomatic patients (such as our three cases), the relative low sensitivity of transthoracic echocardiography (TTE) in discovering CRAT and the spontaneous dissolution of thrombus that has been reported previously [4, 5, 51]. But according to our study, most of cases can be detected by echocardiography. As a result, we thought the echocardiography should be routine screening for hemodialysis patient by the Catheter dialysis with a screening tool for atrial thrombus with sensitivity and specificity of 82.2% and 95.3%, respectively [52]. Magnetic resonance imaging and CT scans could also be used to detect intracardiac thrombi, if echocardiography has not certain in diagnosis of atrial thrombus [48]. Mortality rates was 18.1% in our study. Most of cases can be detected through echocardiography and had no presentation except for infective CRAT. Anticoagulation treatment and surgical thrombectomy were first-line treatment for CRAT in the most cases.

Nowadays, with the extensively use of dialysis catheters, the incidence of CRAT was increasing, but the treatment for this complication has not reached a consensus include pediatric population. As a result, we conducted this study comparing different treatments which the literatures adopted. Firstly, removal or exchange of the hemodialysis catheter appears necessary, as Stavroulopoulos et al. revealed that reserving the catheter was significantly independent factor associated with mortality [3]. Several reasons contributed to such case. Firstly, the mechanism of thrombosis may be the recurrent mechanical stimulation to the atrial wall by the tip of dialysis catheter and the coagulation cascade and platelet aggregation was activated by endothelial damage, causing to the formation of thrombus [17, 53]. Secondly, hemodynamics of the right atrium with the dialysis catheter was relative separation or stasis which may also lead to thrombus formation [22, 54, 55]. As a matter of fact, Gilon et al. revealed that position the tip of catheter in the right atrium is highly correlated with atrial thrombus [54]. This may be the reason that removing the dialysis catheter alone permitted spontaneous dissolution of the thrombosis as reported cases in the literature [51, 56]. The catheter tip of tunnelled cuffed catheters should be in the right atrium according to the recommendation of the National Kidney Foundation: Dialysis Outcomes Quality Initiative guidelines [57]. However, this positioning may increase the incidence of atrial thrombosis and we should balance this relationship. As a result, the prospective studies are needed to performed to validate whether the position of tip of dialysis catheter in the right atrium is an independent risk factor for CRAT or not, and optimize the position of the catheter tip which can not only meet the need of adequate dialysis but also reduce incidence of complications. The presence of bacteraemia is also a reason for removing the catheter because it is hard to solve this problem without catheter removal. According to our analysis, only four cases had not removed the catheter in the patients of infective CRAT, and two of them died. Infection is a common complication of central venous dialysis catheters. It creates an environment of thrombosis formation and it is reversely a nidus for bacteraemia [23, 55]. If vascular access for hemodialysis patients is limited, replacement the dialysis catheter in situ and anticoagulation was also a good choice which was validated by the result of Yang et al. [10] and our cases. However, it was accompanied by a recirculation rate of 17% with good dialysis adequacy because of the step-shaped tip of dialysis catheter. As a result, we attempted to use another kind of dialysis catheter which is the Z-shaped tip in the exchange of catheter so that we can decrease recirculation rate in our three cases. Catheter removal or replacement is the potential risk for pulmonary embolism because of thrombosis movement especially for thrombosis that are large, mobile or adherent to the catheter tip [29, 46]. However, Stavroulopoulos et al. [3] thought that it may be avoidable if we performed catheter manipulation after anticoagulation was in progress.

Removal or replacement of the catheter is critical step for the case of CRAT, but this method alone may be not enough due to no treatment was related with mortality not just for hemodialysis patients [3, 6, 7, 29]. Three strategies were as follows: systemic thrombolysis, anticoagulation and thrombectomy. Cases of thrombolysis were least as reported and only three cases of thrombolysis alone were successful. Most of cases were accompanied thrombolysis with anticoagulation and these cases were all successful, and this method was proved to be a safe and effective approach [48].

Anticoagulants have been recommended as the preferred choice for CRAT if the patients have no contraindications [3, 58]. However, comparing surgical therapy with anticoagulation therapy, no difference had been detected in mortality, although the patients with surgery were younger and more suitable for surgical therapy [3]. This is consistent with other studies with atrial thrombosis for non-hemodialysis patients [11, 31, 55]. Notably, antiplatelet therapy for CRAT is ineffective because the atrial thrombus is a venous thrombosis [46]. In the progress of anticoagulants, serious complications are common such as pulmonary emboli. As a result, the progress should be checked frequently by echocardiography and anticoagulation should be administered for 6 months at least until the thrombosis was dissolved completely, with a target INR 2–3, as this was validated through our analysis. Stavroulopoulos et al. recommended that the patient who has thrombophilia should be lifelong anticoagulation, especially for the patients who continued to be dialyzed through a catheter [3]. If the thrombosis is not dissolved or other serious new complications happened, Surgical thrombectomy should be considered immediately.

If the patients have contraindication for anticoagulation treatment and other serious complications that can be treated simultaneously by the surgery, the patient should be first assessed for thrombectomy [3]. Stavroulopoulos et al. thought surgical thrombectomy should be adopted when the CRAT is ≥ 60 mm [3]. However, we found that it was not successful in all patients with the CRAT ≥ 40 mm by use of anticoagulation. In addition, the presence of patent foramen ovale in the heart may be indication of thrombectomy, because it can extract thrombus and reduce the risk of a subsequent paradoxical embolism, even it can repair it in the surgery. Percutaneous transluminal thrombectomy was an alternative strategy in high-risk patients in whom standard treatment are contraindicated as reported before [59, 60]. However, this method was dependent on the doctor’s experience and have potential complications due to the technical difficulties (the manipulation of the basket in the right atrium is associated with a potential risk of perforation of the atrial wall, or of pulmonary embolism due to clot fragmentation) so that it cannot be routine adopted [18]. Nowadays, no consensus have been rearched about ideal indications for CRAT of hemodiaysis patients.

Our study also has some limitations. Our management method has not been validated through a randomized control trial. In addition, our study has not analyzed risk factors for the development of CRAT. However, this is the largest study for analysis of CRAT in hemodialysis patients and provides advices in the management for it up to now.

Conclusion

Catheter replacement or removal should be the preferred step in the treatment of a CRAT. If the catheter is the only vascular access, we recommended that catheter exchange is a good choice. Combination anticoagulation with thrombolysis may be a main strategy. Surgical thrombectomy is not better than anticoagulation treatment, but it is an alternative strategy under certain conditions. For the cases which have contraindication of anticoagulation and surgery, percutaneous intravascular removal of the thrombosis is an optional method but it needs to be performed by experienced physician. However, prospective studies should be performed for analysis risk factors of development of CRAT in hemodialysis patients in the future.

Author contributions

LC, BC, HG and ZW conceived of the study and participated in its design and coordination. LC and BC participated in the design of the study.GJ, QL, and YZ carried out the selections of patients. WL and SE carried out the measurements of the samples. WL performed the statistical analysis. HG and ZW helped to draft the manuscript. HG and ZW organized the study project and drafted the final manuscript. All authors read and approved the final manuscript. All authors reviewed the manuscript.

Funding

None.

Data availability

The datasets used or analyzed during the current study are available from the corresponding author on reasonable request.

Declarations

Ethics approval and consent to participate

The case was approved by the Human Ethics Committee of the First Affiliated Hospital of Chongqing Medical University (Number:2023 − 315).

Consent for publication

The patient provided written informed consent.

Clinical trial number

Not applicable.

Competing interests

The authors declare no competing interests.