Upper limb deep vein thrombosis (DVT) refers to the formation of a fibrin clot within the subclavian, axillary, and brachial veins of the arm [
All patients hospitalized in the general surgical unit classified as high risk for venous thromboembolism (VTE) with a peripheral intravenous catheter in the dorsal hand or forearm were considered for inclusion. We accepted peripheral intravenous catheters of any gauge. We defined high risk as age over 65 years and one other risk factor either an underlying malignancy, thrombophilia (Factor C deficiency, Factor S deficiency, Factor 5 Leiden mutation, antithrombin 3 deficiency, antiphospholipid syndrome, and prothrombin 20210A mutation), or recent major surgery according to criteria from the Royal Australasian College of Surgeons (RACS) [
The study was a prospective observational single arm cohort study conducted in the general surgical unit of Mackay Base Hospital. The study period occurred from August 2014 to July 2015. The study received ethical approval from the Townsville Hospital and Health Services Human Resources and Ethics Committee on the 30th of May 2014. Informed consent was obtained from all individual participants included in the study. On enrolment, the participant’s demographics, medical comorbidities, full blood count, international normalised ratio of prothrombin time of blood coagulation (INR), and catheter characteristics were collected. All patients had an ultrasound examination of the superficial and deep veins of the cannulated arm on enrolment into the study and at discharge. Colour duplex sonography was selected for diagnosis of DVT as it has a high sensitivity (93%) and specificity (93%) and is noninvasive and easy to perform [
The incidence of peripheral catheter-related DVT has only been reported in one previous study. Therefore we estimated a 20% prevalence based on the results of this study. Based on the estimated prevalence, we calculated a sample size of 62 participants for 95% Confidence Intervals (CI) of 10% to 30%. The incidence of peripheral catheter-related thrombosis is reported as the proportion of enrolled patients who had an ultrasound confirmed thrombus. Corresponding 95% CIs were calculated for the proportions of thrombosis. Descriptive statistics were generated for all included cases. Continuous variables were expressed as medians and interquartile ranges (IQR 25th and 75th percentile) or means and standard deviations (SD). The categorical variables were expressed as numbers and percentages. Bivariate analysis was performed using Fisher’s exact test with thrombosis as the dependent variable and various risk factors as the independent variable. The data was analysed using SPSS version 20 (SPSS Inc., Chicago, IL, USA) for Windows; statistical significance was given to
From August 2014 to July 2015 there were 337 participants aged 65 years and over who had a peripheral catheter placed in the dorsal hand or forearm in the general surgical unit. Two-hundred and sixty patients did not meet eligibility criteria and 17 declined to enter the study. Sixty patients consented to participate in the study. Six patients did not complete the study. Five patients were lost to follow-up and one patient passed away from an underlying medical condition unrelated to the study. In total, data from 54 patients was analysed. The participant characteristics are listed in Table
Characteristics of the participants.
Variable | Result |
---|---|
Age (years), median (IQR) | 73 (69–79.25) |
Gender, |
21 (38.9%) |
Ethnicity, |
|
Caucasian | 53 (98.1%) |
Aboriginal and Torres Strait Islander | 1 (1.98%) |
Anticoagulation, |
25 (46.3%) |
Haemaglobin, median (IQR) | 130 (121.75–142.25) |
Platelets, median (IQR) | 226 (177–316.5) |
INR, median (IQR) | 1.1 (1.0–1.1) |
Catheter size, |
|
18 G | 48 (88.9%) |
20 G | 3 (5.6%) |
22 G | 3 (5.6%) |
Time of catheterization in days, median (IQR) | 3.83 (2–5) |
Location of peripheral catheter, |
|
Dorsum hand | 17 (31.5%) |
Wrist | 6 (11.1%) |
Forearm | 10 (18.5%) |
Cubital fossa | 21 (38.9%) |
Resited peripheral catheters, |
52 (96.3%) |
Number of attempts at catheterization, mean (SD) | 1 (0.79) |
BMI, |
17 (31.5%) |
Diabetes, |
15 (27.8%) |
Malignancy, |
20 (37%) |
Metastatic disease, |
3 (5.6%) |
Recent surgery, |
40 (74.1%) |
Characteristics of the participants with upper extremity thrombosis.
Age | Sex | Location of catheter | Size of catheter | Risk factors | Clinical signs | Type of thrombus | Thrombus location | Size of thrombus | Treatment | Symptoms at 30 days |
---|---|---|---|---|---|---|---|---|---|---|
71 | F | L forearm 5 cm from wrist | 20 G | Recent surgery, type 2 DM, and obesity | No signs | SVT | Cephalic vein → 5 cm from cannula site | Speck | LMWH | No symptoms |
|
||||||||||
85 | F | L cubital fossa | 20 G | Colorectal cancer, recent surgery | No signs | SVT | Cephalic vein | 7 cm segment of thrombus | LMWH | No symptoms |
|
||||||||||
72 | F | L cubital fossa | 20 G | Bladder lymphoma | No signs | SVT | Cephalic vein | Thrombus at site of cannula | LMWH | No symptoms |
|
||||||||||
81 | F | L cubital fossa | 20 G | Colorectal cancer, recent surgery | No signs | SVT | Cephalic vein | Segment of thrombus | LMWH | No symptoms |
|
||||||||||
83 | M | R dorsal hand | 22 G | Recent surgery | No signs | DVT | Brachial vein | Segment of thrombus | Warfarin | No symptoms |
|
||||||||||
70 | M | L cubital fossa | 20 G | Recent surgery | No signs | SVT | Basilic vein | Nonocclusive | Warfarin | No symptoms |
Risk factors.
Risk factor | Number of thromboses |
|
---|---|---|
Gender (female) | 4/21 (19%) | 0.193 |
Ethnicity (Caucasian) | 6/53 (11.3%) | 1 |
Diabetes | 1/15 (6.66%) | 1 |
Metastatic disease | 0/3 (0%) | 1 |
Anticoagulation | 3/25 (12%) | 1 |
Elevated WCC | 1/8 (12.5%) | 1 |
BMI >30 | 2/17 (11.76%) | 1 |
Cannula LOS >1 | NS | |
Size of cannula | NS | |
Cubital fossa | 4/21 (19%) | NS |
More than 1 attempt | 1/12 (8.33%) | NS |
Malignancy | 3/20 (15%) | 0.659 |
Surgery | 4/40 (10%) | 0.643 |
Ours is the first prospective study designed to detect the incidence of peripheral catheter-related DVT and to identify associated risk factors in surgical patients. We detected a low incidence of peripheral catheter-DVT (incidence of 1.8%) with no associated risk factors. However there was a higher incidence of peripheral catheter-related SVT (9.8%). All cases of peripheral catheter-related thrombosis in our study were asymptomatic and detected on ultrasound as a result of our study. There were no cases of pulmonary embolism clinically detected in our study.
The low incidence of peripheral catheter-related DVT reinforces the results of a previous study. A randomised control trial comparing the efficacy of PICCS and peripheral catheters found a 3.4% incidence of peripheral catheter-related DVT. Similar to our study ultrasound was performed at the beginning and end of catheterization. There was a higher incidence of peripheral catheter-related SVT than the incidence of peripheral catheter-related DVT in our study. The incidence of peripheral catheter-related SVT was lower than figures reported in a previous study (9.2% compared to 44%). The increased incidence of SVT in Periard’s study is likely attributed to larger gauge of catheters used and increased time of catheterization [
We assessed recent surgery, malignancy, metastatic disease, increased BMI, diabetes mellitus, cannula gauge, number of attempts at cannulation, and length of catheterization for association with catheter-related thrombosis. None of the risk factors assessed in our study were found to have a significant association with catheter-related thrombosis. However, the data analysis was limited by the small number of events. In our study 83% of participants with upper extremity thrombosis had recent surgery. Previous studies report mixed results regarding recent surgery as a risk factor for catheter-related thrombosis [
The results of our study raise questions regarding screening for asymptomatic upper extremity thrombosis. None of the patients with peripheral catheter-related thrombosis were symptomatic. Previous studies have reported similar findings where all cases of thrombosis were found incidentally on ultrasound [
There are strong arguments in literature regarding anticoagulation of symptomatic upper limb DVT as it prevents propagation of the thrombus and development of a pulmonary embolism (PE) [
We detected five cases of SVT in our study. SVT has previously been regarded as a benign disease; however there is a growing body of evidence to suggest otherwise [
There were no cases of clinically detected PE in our study. Similar findings have been reported in previous studies. In a large retrospective series, Chemaly et al. reported an incidence of 3.4% of PE in patients with PICC-lines [
There are several limitations to our study. The present study is a prospective observational study conducted in a single regional institution; therefore the findings of this study may not be applicable to other hospital populations. There is inherent bias associated with the observational design of the study as confounders are unable to be controlled and there are a lack of internal controls. The study was underpowered; therefore the confidence intervals are very broad, indicating that the results are imprecise. The low incidence of peripheral catheter-related thrombosis limited the ability to identify associated risk factors. Nonconsecutive patients were enrolled into the study; therefore the study sample may not be a true reflection of the intended sample. Ideally, both arms would be scanned for a DVT; however this would have been too difficult considering financial constraints of the study. Furthermore patients were only followed up for 30 days; therefore symptoms of upper limb DVT after 30 days are unknown.
Despite the limitations, there were also several strengths with this study. This is the first study to investigate the incidence of peripheral catheter-related venous thrombosis in surgical patients. The prospective design and investigation of all study participants prevented missing any case of peripheral catheter-related thrombosis. In addition this confirmed the asymptomatic nature of peripheral catheter-related thrombosis. We also used objective investigations to diagnose thrombosis preventing misdiagnosis. This is also the first study to investigate the risk factors for upper limb superficial venous thrombosis.
In our study we recorded a low incidence of peripheral catheter-related DVT (1.8%, 95% CI −1.7% to 5.4%), but a higher incidence of superficial venous thrombosis (9.25%, 95% CI 1.5% to 17%). The small sample size and low incidence of peripheral catheter-related thrombosis limited our ability to determine the association between peripheral catheters and upper limb DVT. Future studies will be required to show the natural history of an asymptomatic upper limb DVT through regular ultrasounds and investigation for occult pulmonary emboli.
All procedures performed in studies involving human participants were in accordance with the ethical standards of the institutional and/or national research committee and with the 1964 Helsinki declaration and its later amendments or comparable ethical standards.
None of the authors have any conflict of interests.
The authors would like to acknowledge the assistance of Dr. Antonio Pace for reporting the ultrasounds and the sonographers at Mackay Base Hospital (Susanna Rae, Ainslie Heinke, Heidi Croxson and Amy Leigh). Funding for the project was obtained from the Private Practice Trust Fund and James Cook University School of Medicine and Dentistry.