Catheter-related Thrombosis in Patients with Lymphoma or Myeloma Undergoing Autologous Stem Cell Transplantation

Abstract

Catheter-related thrombosis (CRT) occurs frequently during autologous hematopoietic cell transplantation (AHCT) and data regarding the incidence, risk factors, and management are understudied. We evaluated 789 consecutive patients with lymphoma and myeloma that underwent AHCT over 10 years (2006 to 2016) and detected the incidence of CRT was 6.3%; only 32% of CRT were symptomatic. The majority occurred within 100 days of AHCT (86%) and median time from tunneled line placement to CRT was 44 days (11-89). Outcomes of these 50 patients with CRT were compared with age and disease-matched AHCT controls to identify risk factors. History of prior venous thromboembolism (VTE, 20.9% vs 7.0%, p=0.02) was the only significant risk factor. Treatment with low-molecular weight heparin was tolerated with rare minor bleeding (4%), although CRT recurrence/extension (10%) and subsequent VTE (12%) were common. CRT did not impact on non-relapse mortality (NRM) or risk of relapse; 2-year progression-free survival was 55% in CRT cases vs. 54% in controls (p=0.42). CRT appears to be common in patients with lymphoma and myeloma undergoing AHCT and significantly contributes to morbidity. Further study to determine mitigating strategies and modiors for CRT is warranted.

Introduction

Use of central venous catheters (CVCs) for cell apheresis, medications, blood products and total parenteral nutrition is integral to the process of autologous hematopoietic cell transplantation (AHCT) of hematologic malignancies. Large bore (14-15 french) CVCs are routinely used to collect hematopoietic stem cells and CVC insertion is complicated by catheter-related thrombosis (CRT) in up to 20% of patients.^1-3 Policies on how long to keep the catheters in place and timing of catheter removal is highly variable and practice management guidelines are inconsistent. CRT represent a major complication of indwelling CVCs, yet data on incidence, timing in relationship to line insertion and removal, and subsequent complications and impact of CRT on morbidity of AHCT are often overlooked and understudied.

The increased incidence of thrombosis in peri-transplant setting can be attributed to endothelial damage by conditioning regimens, systemic infections including bacteremia and transient immobilization in this population.⁴ Alterations in coagulation proteins are also thought to contribute, such as increase in von Willebrand factor, platelet adhesion and thrombin generation with decrease in protein C levels.^5-7 Standard treatment guidelines are not available for the AHCT population with special concerns such as prolonged severe thrombocytopenia, platelet tranfusion dependence, and ongoing need for catheter placement.^{1, 8} Furthermore, whether prompt CVC removal and anticoagulation prophylaxis might reduce the incidence of CRT in transplant patients has not been studied systematically. This study aimed to evaluate the incidence, risk factors, management patterns and outcomes of CRT in patients with lymphoma or myeloma undergoing AHCT in our transplantation center.

Patients and Methods

Using prospectively collected data from University of Minnesota Blood and Marrow Transplantation Database, we studied 789 consecutive patients with lymphoma or multiple myeloma who underwent AHCT between 2006-2016. Available medical records were supplemented by chart review for the occurrence of CRT. We identified 50 subjects with CRT from CVCs. Imaging was performed for symptoms or signs of venous thrombosis (pain, swelling, catheter malfunction) while asymptomatic CRT events were detected on routine surveillance imaging or imaging done to evaluate an unrelated symptom. CRT was defined as mural thrombus with partial or total occlusion of a vessel, pulmonary embolism or right atrial thrombus in which a catheter was present or had been present within the prior 30 days. Fibrin sheaths were not included. Patient and transplant characteristics and outcomes were cored in the case-control analysis with 2:1 matching by patient age and disease of AHCT recipients. The University of Minnesota Institutional Review Board approved the study.

Transplant procedures

The study eligibility criteria included adults (>18 years) with relapsed non-Hodgkin NHL), Hodgkin lymphoma (HL), and multiple myeloma (MM) with chemo-sensitive disease and no extreme organ dysfunction.⁹ All AHCT recipients underwent insertion of tunneled 14.5fr double lumen large bore tunneled catheters via left or right subclavian access used for hematopoietic cell apheresis between 30 to 7 days prior to start of conditioning regimen. Some patients had prior port-a-cath placed or had peripherally inserted central catheter (PICC) or tunneled central line for administration of salvage anti-neoplastic chemotherapy. Transplant conditioning consisted of melphalan 200 mg/m² for MM. The BEAM conditioning regimen consisted of carmustine 300 mg/m² IV once on day −6, etoposide 100 mg/m² IV twice daily on days −5 to −2, cytarabine 100 mg/m² IV twice daily on days −5 to −2, and melphalan mg/m² once on day −1. Some patients received cyclophosphamide 60 mg/m² IV for 2 days plus total body irradiation (TBI) 1320 cGy or CBV (cyclophosphamide 1.5 g/m² daily for 4 days, carmustine 300 mg/m² for 1 day, and etoposide 150 mg/m² IV for 6 doses). G-CSF started on Day +5. Venous Doppler ultrasound were done if patients developed symptoms or signs of venous thrombosis (upper extremity swelling, neck pain/swelling, or pain at insertion site of along tunnel) or catheter malfunction.

Study objectives and statistical methods

The primary objective was to describe the CRT incidence, management and complications. Secondary objective was to identify CRT risk factors using case-control cohorts. Patients and disease characteristics were summarized using descriptive statistics. Statistical comparison of categorical variables was performed by χ2 and Wilcoxon rank-sum test was used for assessment of continuous variables between cases and controls. The Kaplan-Meier method¹⁰ was used to estimate the probabilities of PFS, and the log-test was used for univariate comparisons between groups. The cut-off significance level for all P values was 0.05. All statistical analyses were performed with Statistical Analysis System statistical software version 9.3 (SAS Institute, Inc., Cary, NC) and R Statistical Software (Foundation for Statistical Computing, Vienna, Austria, http://www.R-project.org).

Results

Patient and CRT cases characteristics

The incidence of CRT in patients with lymphoma and multiple myeloma undergoing AHCT over the ten-year period was 6.3% (n=50). Most were detected incidentally but 32% were symptomatic from the thrombosis. Five of 50 patients had pulmonary embolism detected incidentally by CT imaging; all were asymptomatic. Of the 50 patients with CRT, 22% had refractory Hodgkin’s lymphoma (HL; incidence in HL 11.0%), 62% non-Hodgkin’s lymphoma (NHL; incidence in NHL 10.1%) and 16% myeloma (MM; incidence in MM 2.1%). The median age at the time of CRT was 56.6 years; 60% were male.

The main demographic summary of CRT cases and catheter characteristics are shown in Table 1. Out of 50 cases of CRT, 43 occurred within the acute post-transplant phase of 100 days after AHCT. The majority (94%) of cases were associated with tunneled line that were double lumen (90%), while 4% were related to PICC. The CVC insertion site was predominantly internal jugular (90%). Half of CRT cases had concomitant Port-a-Cath and 46% had more than one CVC inserted peritransplant. Vein occlusion was complete in 46% and occurred with the similar frequency on left and right sides (44% vs 42%). In the patients with CRT within 100 days of autologous transplant, the median time from tunneled line placement to CRT was 44 days (range, 11-89). Majority of patients had tunneled line removed on same day as CRT was identified, and median time from CVC insertion to removal was 42 days (range 10-73 days).

Table 1.

Patient, catheter, and transplant characteristics. Catheter-related thrombosis (CRT) cases are divided into those occurring within first 100 days post-transplant (n=43) versus beyond 100 days (n=7).

Characteristics	All Cases (n=50)	CRT beyond 100 days of AHCT (n=7)	CRT within 100 days of AHCT (n=43)	P-value
Recipient gender, male, n (%)	30 (60.0%)	4 (57.1%)	26 (60.5%)	0.87
Age at AHCT, median (range), yr	54.3 (19.7-70.9)	50.8 (28.9-68.8)	57.1 (19.7-70.9)	0.41
Age at CRT, median (range), yr	56.6 (19.7-71.0)	56.4 (30.7-69.3)	56.9 (19.7-71.0)	0.75
Second transplant after AHCT, n (%)	10 (20.0%)	5 (71.4%)	5 (11.6%)	<0.01
Preparative Regimen, n (%)				0.23
BEAM	16 (32.0%)	0	16 (37.2%)
CBV	13 (26.0%)	3 (42.9%)	10 (23.3%)
CyTBI	13 (26.0%)	2 (28.6%)	11 (25.6%)
Melphalan	8 (16.0%)	2 (28.6%)	6 (14.0%)
Karnofsky score group, n (%)				0.28
Missing	5 (10.0%)	1 (14.3%)	4 (9.3%)
<90	8 (16.0%)	2 (28.6%)	6 (14.0%)
≥90	37 (74.0%)	4 (57.1%)	33 (76.7%)
HCT-CI, n (%)				0.64
0	26 (52.0%)	4 (57.1%)	22 (51.2%)
1-2	19 (38.0%)	3 (42.9%)	16 (37.2%)
>=3	5 (10.0%)	0	5 (11.6%)
Port-a-cath, n (%)	25 (50.0%)	5 (71.4%)	20 (46.5%)	0.22
Port-a-cath removed, n (%)	4 (8.0%)	1 (14.3%)	3 (7.0%)	0.79
Time from CRT to CVC removal
Median (range), days	0 (−21-161)	104 (46-161)	0 (−21-26)	0.02
Missing	6	5	1
Symptomatic CRT, n (%)	16 (32.0%)	4 (57.1%)	12 (27.9%)
CVC Type, n (%)				0.12
Tunneled CVC only	24 (48.0%)	2 (28.6%)	22 (51.2%)
PICC only	2 (4.0%)	1 (14.3%)	1 (2.3%)
Port only	1 (2.0%)	0	1 (2.3%)
Multiple central lines	23 (46.0%)	4 (57.1%)	19 (44.2%)
CVC caliber				0.30
CVC caliber single	1(2.0%)	0	1(2.3%)
double	26(52.0%)	3(42.9%)	23(53.5%)
double plus single	18(36.0%)	2(28.6%)	16(37.2%)
double double	1(2.0%)	0	1(2.3%)
Number of lines on day of CRT, n (%)				0.54
1	40 (80.0%)	5 (71.4%)	35 (81.4%)
2	10 (20.0%)	2 (28.6%)	8 (18.6%)
CVC location, n (%)				0.92
R internal jugular	25 (50.0%)	3 (42.9%)	22 (51.2%)
L internal jugular	20 (40.0%)	2 (28.6%)	18 (41.9%)
R subclavian	1 (2.0%)	0	1 (2.3%)
Site of Occlusion, n (%)				0.14
Right neck	21 (42.0%)	3 (42.9%)	18 (41.9%)
Left neck	22 (44.0%)	2 (28.6%)	20 (46.5%)
Atrium	1 (2.0%)	0	1 (2.3%)
Pulmonary embolism	5 (10.0%)	1 (14.3%)	4 (9.3%)
Inteferior Vena Cava	1 (2.0%)	1 (14.3%)	0
Amount of Occlusion, n (%)				0.48
Complete	23 (46.0%)	4 (57.1%)	19 (44.2%)
Partial	17 (34.0%)	1 (14.3%)	16 (37.2%)
Recurrent CRT, n (%)	5 (10.0%)	0	5 (11.6%)	0.34
VTE other than CRT, n (%)	14 (28.0%)	5 (71.4%)	9 (20.9%)	<0.01
VTE prior to CRT, n (%)	8 (16.0%)	4 (57.1%)	4 (9.3%)	<0.01
VTE after CRT, n (%)	6 (12.0%)	1 (14.3%)	5 (11.6%)	0.84
Prior radiotherapy, n (%)	9 (18.0%)	1 (14.3%)	8 (18.6%)	0.78
Bulky mediastinal mass, n (%)	6 (12.0%)	1 (14.3%)	5 (11.6%)	0.84
Prior splenectomy, n (%)	3 (6.0%)	1 (14.3%)	2 (4.7%)	0.32
WBC peak at stem cell collection				0.14
Median (range), ×109/L	42.9 (2.2-96.1)	29.0 (27.1-30.9)	43.5 (2.2-96.1)
WBC peak at engraftment				0.07
Median (range), ×109/L	14.5 (1.4-30.9)	7.8 (1.4-20.0)	15.2 (4.6-30.9)
Platelet count at CRT				0.02
Median (range), ×109/L	123 (29-426)	69 (29-124)	133 (31-426)
INR at CRT				0.66
Median (range), seconds	1.0 (0.9-1.3)	1.1 (0.9-1.3)	1.0 (0.9-1.2)
PTT at CRT				0.50
Median (range), seconds	33.5 (25.0-66.0)	28.0 (26.0-38.0)	34.0 (25.0-66.0)
Fibrinogen at CRT				0.28
Median (range), mg/dL	466 (402-622)	433 (402-546)	493 (439-622)
Time to >50,000 platelet recovery				0.84
Median (range), days	15 (9-36)	15 (11-28)	15 (9-36)
Received lytic therapy, n (%)	1 (2.0%)	0	1 (2.3%)	0.68
Bleeding complication, n (%)	2 (4.0%)	0	2 (4.7%)	0.56
CVC bactermia, n (%)	9 (18.0%)	2 (28.6%)	7 (16.3%)	0.43
Mucositis, n (%)	28 (56.0%)	3 (42.9%)	25 (58.1%)	0.45
Number platelet transfusions in 14 days post-AHCT				0.06
Median (range)	4 (0-12)	2 (1-7)	4 (0-12)
Received TPN, n (%)	7 (14.0%)	1 (14.3)	6 (14.0%)	0.98
Received hormonal therapy, n (%)	4 (8.0%)	0	4 (9.3%)	0.40
Anticoagulation, n (%)				0.09
Enoxaparin	15 (30.0%)	2 (28.6%)	13 (30.2%)
Warfarin	28 (56.0%)	3 (42.9%)	25 (58.1%)
Unfractionated heparin	1 (2.0%)	1 (14.3%)	0
None	6 (12.0%)	1 (14.3%)	5 (11.6%)
Treatment duration, n (%)				0.95
≤3 month	19 (38.0%)	3 (42.9%)	16 (37.2%)
3 to 6 months	12 (24.0%)	2 (28.6%)	10 (23.3%)
≥6 months	11 (22.0%)	1 (14.3%)	10 (23.3%)
None	6 (12.0%)	1 (14.3%)	5 (11.6%)
Prophylactic anticoagulation after AHCT, n (%)	2 (4.0%)	1 (14.3%)	1 (2.3%)	0.24

Abbreviations: AHCT = autologous hematopoeitic cell transplantation; CRT = catheter-related thrombosis; CVC = central venous catheter; BEAM = carmustine, etoposide, cytarabine, melphalan; CBV = cyclophosphamide, etoposide, carmustine; CyTBI = cyclophpsphamide, total body irradiation; VTE = venous thromboembolism; TPN = total parenteral nutrition, WBC = white blood cells; INR = international normalized ratio; PTT = partial thromboplastin time.

History of radiation to chest/neck (18%), prior bulky mediastinal mass (12%), prior splenectomy (6%), post-AHCTbacteremia (18%), mucositis post-AHCT (56%), TPN (14%) post-AHCT, and hormonal therapy peri-transplant (8%) were cor amongst CRT cases. Median international normalized ratio (INR, 1.0, <1.2 seconds) and partial thromboplastin time (PTT, 33.5, range 25-35 seconds) were were normal and fibrinogen was elevated in most patients (median 466, range 402-622 mg/dL) at time of CRT.

CRT management

Most patients (86%) with CRTs were treated with low molecular weight heparin (LMWH, 86%) and 56% received subsequent warfarin. Only 2% received unfractionated heparin. LMWH at dose 1 mg/kg twice daily was administered if platelet count could be maintained above 50 × 10⁹/L and median time to platelet recovery in CRT cases was 15 days (range 9-36). At the time of CRT, median platelet count was 133 (range 31-426 × 10⁹/L). Seven (14%) of patients developed CRT before platelet count reached 50 × 10⁹/L. Patients with CRT received median of 4 units of irradiated platelet transfusions (range 0-12) post-transplant. Patients were anticoagulated for <3 months (38%), 3-6 months (24%), and >6 months (22%). One patient required lytic therapy for CRT. Only two patients (4%) received secondary prophylactic anticoagulation after AHCT. This management was safe and effective and grade 1-2 bleeding events occurred in only two patients (4%) during anticoagulation. Five patients (10%) had CRT recurrence or extension while 12% had subsequent VTE after CRT.

Risk factors for CRT and transplant outcomes

To examine CRT risk factors, we compared 43 CRT cases occurring within first 100 days post-transplant with 86 age and disease-matched controls undergoing AHCT (at ratio 1:2) in Table 2. The main risk factors for CRT included prior VTE event (20.9% vs 7.0%, p=0.02), higher white blood cell (WBC) peak at mobilization (43.5 vs 40.4 × 10⁹/L, p=0.17) and peak WBC at engraftment (15.2 vs 11.9 × 10⁹/L, p=0.18). Gender, age, presence of Port-a-Cath, hematopoietic cell transplantation comorbidity-index¹¹ (HCT-CI), and Karnofsky score did not increase the risk of CRT. Two patients underwent a second autologous and seven patients an allogeneic stem cell transplant. No deaths occurred due to CRT or its treatment. Two-year progression-free survival after AHCT was 54% (95% CI 50-57%) and was not affected by CRT (55% vs 54%, p=0.42) as shown in Figure 1. NRM and relapse rates were similar by CRT (data not shown).

Table 2.

Univariate Analysis of CRT Cases vs Controls without CRT. Catheter-related thrombosis (CRT) occurring within first 100 days post-transplant (n=43) are compared 2:1 to age and disease-matched controls (n=86).

	All Groups (n=129)	CRT within 100 days of AHCT (n=43)	Controls without CRT (n=86)	P-value
Recipient gender, n (%)				0.70
Male	81 (62.8%)	26 (60.5%)	55 (64.0%)
Female	48 (37.2%)	17 (39.5%)	31 (36.0%)
Age at AHCT				0.96
Median(Min- Max)	57.0 (19.6-70.9)	57.1 (19.7-70.9)	56.9 (19.6-70.7)
Age Cutoff, n (%)				1.00
<50 years	36 (27.9%)	12 (27.9%)	24 (27.9%)
≥50 years	93 (72.1%)	31 (72.1%)	62 (72.1%)
Peak WBC at Mobilization				0.17
Median (range), ×109/L	42.1 (2.2-105.9)	43.5 (2.2-96.1)	40.4 (17.0-105.9)
Peak WBC at Engraftment				0.18
Median (range), ×109/L	12.5 (3.8-52.8)	15.2 (4.6-30.9)	11.9 (3.8-52.8)
Port-a-cath, n (%)				1.00
No	69 (53.5%)	23 (53.5%)	46 (53.5%)
Yes	60 (46.5%)	20 (46.5%)	40 (46.5%)
History of VTE, n (%)				0.02
No	114 (88.4%)	34 (79.1%)	80 (93.0%)
Yes	15 (11.6%)	9 (20.9%)	6 (7.0%)
HCT-CI, n (%)				0.13
0-2	88 (68.2%)	38 (88.4%)	50 (58.1%)
>=3	20 (15.5%)	5 (11.6%)	15 (17.4%)
Missing	21 (16.3%)	0	21 (24.4%)
Karnofsy Score, n (%)				0.37
<90	14 (10.9%)	6 (14.0%)	8 (9.3%)
>=90	107 (82.9%)	33 (76.7%)	74 (86.0%)
Missing	8 (6.2%)	4 (9.3%)	4 (4.7%)

Abbreviations: AHCT = autologous hematopoeitic cell transplantation; CRT = catheter-related thrombosis; WBC = white blood cells; VTE = venous thromboembolism; HCT-CI = hematopoetic cell transplantation comorbidity-index.

Figure 1.

Two-year Progression Free Survival by CRT Group. PFS for all patients was 54% and was not affected by CRT (55% vs 54%, p=0.42).

Discussion

In patients undergoing AHCT, CVC is not only required to secure peripheral blood hematopoietic cell graft but central venous access is integral for preparation regimens, transfusions, and parenteral support needed for care during peri-transplant period. Here, we report our observations on thromboembolic complications associated with central lines. The incidence of CRT of 6.3% for patients with lymphoma and multiple myeloma undergoing AHCT are similar with those reported previously.^{1-3, 12-15} In the largest, retrospective study of thrombotic complications in stem cell transplant, Gerber et al. reported incidence of venous thromboembolism of 4.9%, majority of which were catheter-related.¹³ Yeral et al. showed that symptomatic CRT was 2.7% in their transplant population¹², which corresponds to the one third of our CRT cases that were symptomatic (incidence 2.0%). The remaining asymptomatic CRT events were detected on routine surveillance imaging or venographic studies done to evaluate catheter malfunction. In literature, it is unclear whether the reported 24-70% of asymptomatic catheter-related thrombosis eventually become clinically symptomatic^{1, 2, 16, 17}

The diagnosis of CRT appears to be more common in patients with Hodgkin’s lymphoma than non-Hodgkin’s lymphoma or multiple myeloma which is consistent with findings by Conlan et al.¹ It is speculated that the increased risk in Hodgkin’s lymphoma is secondary to mediastinal site or disease or prior radiation to the area involved by thrombosis.¹ Indeed the majority of patients with neck and mediastinal radiation had Hodgkin’ lymphoma. The risk of thrombosis in multiple myeloma is low despite the use of lenalidomide after AHCT which may have added thrombotic risk.⁸ The need for prophylactic anticoagulation therapy has been based on medical history and most patients receive aspirin or LMWH prophylaxis.⁸ We showed that the majority of CRT events occurred at median 4 weeks post-transplant period which is consistent with Gerber et al. report of thromboembolic events occurring at median interval of 35 days from either autologous or allogeneic transplantation.¹³ For the 7 cases of CRT that occurred greater then 100 days after AHCT, 5 occurred in relation to a subsequent allogeneic transplant suggesting that stem cell transplantation procedure itself is a highly pro-thrombotic state.

Risk factors related to the underlying hematologic malignancy, stem cell transplantation or catheter characteristics may contribute to development of CRT. Our data suggest that higher WBC peak at mobilization and stem cell engraftment could increase the risk for thrombosis, although we lacked statistical significance. The connection between leukocytosis, particularly neutrophilia, and thrombosis in cancer patients is under active investigation ^{19, 20 21, 22} For example Khorana risk score for VTE in cancer patients was developed to predict the risk for cancer patients depending of type of cancer.²³ Using the predictive model, diagnosis of Hodgkin’s lymphoma, elevated prechemotherapy platelet count, and leukocyte count above 11 × 10/L each independently increase VTE risk.²⁴ We also evaluated basic coagulation parameters and confirmed that most patients have normal baseline levels of INR and PTT, and mildly elevated fibrinogen, reflecting hospitalized population.²⁵ Conlan et al. and others had reported low antithrombin III and protein C levels in AHCT patients ^{1, 26}, while increased von Willebrand factor and platelet adhesion may also play a role.⁶ Presence of inherited thrombophilia was rare and its contribution is difficult to assess.⁴ We found that double lumen catheters in the internal jugular vein were most commonly associated with thrombosis then single lumen or subclavian vein catheters, suggesting more caution is needed in some CVC approaches. Previous VTE is a well-recognized risk factor for subsequent VTE^{8, 27} and we show this holds true for CRT as well. While we identified number of risk factors for post-AHCT VTE, we also acknowledge the retrospective nature of this study, incomplete list of potential risk factors analyzed and inherent bias of case-control analysis.

Management of CRT remains highly variable. There are currently no controlled studies of empiric prophylaxis in AHCT recipients and too few patient patients received prophylaxis in this study to draw conclusions. A retrospective study by Lagro et al. reported no benefit for a 7-10 day course of LMWH following catheter placement in VTE reduction.²⁸ Studies in cancer patients not specific to hematologic malignancy, found that longer 6-week course of LMWH or warfarin after CVC insertion did not reduce the rate of CRT.^{29, 30} However, in AHCT setting, some patients with high risk clinical factors, such as large-bore CVC and previous VTE should be considered for secondary prophylaxis with LMWH and this strategy warrants further study. For acute CRT, anticoagulation is recommended regardless of whether the catheter is removed to reduce symptoms, maintain venous access, and prevent chronic venous stenosis or clot extension.^{8, 31-33} Catheter removal is not absolutely necessary if symptoms improve.^{8, 31} We showed that LMWH was well-tolerated with minimal bleeding complications despite concerns given to increased bleeding risk from mucositis and thrombocytopenia posttransplant. In fact, almost all patients with CRT manifested the median platelet count which was acceptable for therapeutic anticoagulation (median greater than 100× 10⁹/L) and anticoagulation appeared safe with no significant bleeding episodes. On contrary, our data suggest that duration of anticoagulation for CVC associated CRT must be individualized as at least one in ten patients had recurrent VTE patients. Given the American College of Chest Physicians (ACCP) guidelines recommended 3 month minimum treated for CRT, about a third of patients in our cohort received short term anticoagulation.³⁴ Duration of anticoagulation after AHCT is therefore an area which needs further exploration. Randomized controlled study of thromboprophylaxis and therapeutic management of CRT is needed to clarify the risks and benefits in AHCT recipients.

In conclusion, the diagnosis of CRT frequently complicates AHCT and appears to be more common in patients with lymphoma than myeloma. In addition to prior VTE which is the main risk factor for CRT, increased risk is observed with double lumen subclavian insertion approach, at post-engraftment time period and with elevated WBC count. Prophylactic anticoagulation to prevent CRT is not routinely recommended but should be considered in patients with history of VTE. Close monitoring and early surveillance is recommended in patients with other risk factors and removal of large-bore CVC early post-apheresis also could reduce the risk of CRT associated morbidity. Therapeutic anticoagulation with LMWH or warfarin appears safe, yet recurrent VTE events highlight the need to individualize duration of anticoagulation. Consideration for secondary prophylaxis should be given to patients with history of CRT undergoing second transplant or other high pro-thrombotic state procedures. Tranplantation centers should implement a CVC guideline and active quality improvement approach to decrease the burden of CRT complications.³⁵ Further prospective study to define additional risk factors and examine the prophylactic and therapeutic management strategies is warranted.

Highlights.

• Catheter-related thrombosis frequently complicates autologous hematopoietic cell transplantation

• History of prior venous thromboembolism is a significant risk factor.

• Catheter-related thrombosis appears to be more common in patients with lymphoma then myeloma undergoing autologous transplantation

• Elevated white blood cell count at stem cell mobilization and engraftment may be a risk factor