If you don't remember your password, you can reset it by entering your email address and clicking the Reset Password button. You will then receive an email that contains a secure link for resetting your password
If the address matches a valid account an email will be sent to __email__ with instructions for resetting your password
The Michigan Hospital Medicine Safety Consortium, Ann Arbor, MichDepartment of Internal Medicine, Division of General Medicine, University of Michigan School of Medicine, Ann Arbor, MichVA Ann Arbor Health Care System, Ann Arbor, Mich
The Michigan Hospital Medicine Safety Consortium, Ann Arbor, MichDepartment of Internal Medicine, Division of General Medicine, University of Michigan School of Medicine, Ann Arbor, Mich
The Michigan Hospital Medicine Safety Consortium, Ann Arbor, MichDepartment of Internal Medicine, Division of General Medicine, University of Michigan School of Medicine, Ann Arbor, MichVA Ann Arbor Health Care System, Ann Arbor, Mich
Requests for reprints should be addressed to Vineet Chopra, MD, MSc, University of Michigan, North Campus Research Complex, 2800 Plymouth Rd, Building 16, #432W, Ann Arbor, MI 48109.
The Michigan Hospital Medicine Safety Consortium, Ann Arbor, MichDepartment of Internal Medicine, Division of General Medicine, University of Michigan School of Medicine, Ann Arbor, MichVA Ann Arbor Health Care System, Ann Arbor, Mich
Peripherally inserted central catheters are associated with upper-extremity deep vein thrombosis. Whether they also are associated with lower-extremity deep vein thrombosis or pulmonary embolism is unknown. We examined the risk of venous thromboembolism in deep veins of the arm, leg, and chest after peripherally inserted central catheter placement.
Methods
We conducted a multicenter, retrospective cohort study of 76,242 hospitalized medical patients from 48 Michigan hospitals. Peripherally inserted central catheter presence, comorbidities, venous thrombosis risk factors, and thrombotic events within 90 days from hospital admission were ascertained by phone and record review. Cox proportional hazards models were fit to examine the association between peripherally inserted central catheter placement and 90-day hazard of upper- and lower-extremity deep vein thrombosis or pulmonary embolism, adjusting for patient-level characteristics and natural clustering within hospitals.
Results
A total of 3790 patients received a peripherally inserted central catheter during hospitalization. From hospital admission to 90 days, 876 thromboembolic events (208 upper-extremity deep vein thromboses, 372 lower-extremity deep vein thromboses, and 296 pulmonary emboli) were identified. After risk adjustment, peripherally inserted central catheter use was independently associated with all-cause venous thromboembolism (hazard ratio [HR], 3.16; 95% confidence interval [CI], 2.59-3.85), upper-extremity deep vein thrombosis (HR, 10.49; 95% CI, 7.79-14.11), and lower-extremity deep vein thrombosis (HR, 1.48; 95% CI, 1.02-2.15). Peripherally inserted central catheter use was not associated with pulmonary embolism (HR, 1.34; 95% CI, 0.86-2.06). Results were robust to sensitivity analyses incorporating receipt of pharmacologic prophylaxis during hospitalization.
Conclusions
Peripherally inserted central catheter use is associated with upper- and lower-extremity deep vein thrombosis. Weighing the thrombotic risks conferred by peripherally inserted central catheters against clinical benefits seems necessary.
This analysis of more than 70,000 hospitalized general medical patients reports that peripherally inserted central catheter insertion is independently associated with both upper- and lower-extremity deep vein thrombosis, even after adjustment for multiple putative risk factors.
•
Receipt of pharmacologic deep vein thrombosis prophylaxis was not associated with a decrease in risk of thrombotic events in upper or lower extremities.
•
Although the mechanisms underlying the association between peripherally inserted central catheter receipt and lower-extremity thrombosis are unclear, the risk of thrombosis in multiple locations stemming from peripherally inserted central catheter use should be weighed carefully against clinical benefits for each patient.
Compared with central venous catheters, use of peripherally inserted central catheters has grown substantially in hospitalized patients.
Increasing use of peripherally inserted central catheters reflects their clinical advantages. For example, peripherally inserted central catheters avoid iatrogenic complications often associated with placement of central venous catheters in the neck or chest. Because peripherally inserted central catheters terminate in central veins, they can be used for tasks as diverse as infusion of antibiotics, chemotherapy, or hemodynamic monitoring. With the advent of hospital-based vascular nursing teams who provide cost-effective insertion at the bedside, peripherally inserted central catheters also have become increasingly accessible and convenient for medical providers.
These advantages notwithstanding, accumulating evidence suggests that peripherally inserted central catheters are associated with important complications, including upper-extremity deep vein thrombosis and pulmonary embolism.
From a physiologic perspective, these adverse events are not surprising. Peripherally inserted central catheters occupy much of the cross-sectional diameter of peripheral veins of the arm and thus predispose to venous stasis.
Finally, because a prime indication for peripherally inserted central catheters is infusion of chemotherapy in patients with cancer, peripherally inserted central catheters often satisfy Virchow's triad, creating the perfect milieu for thrombosis.
existing risk prediction tools and evidence-based guidelines include consideration of peripherally inserted central catheters and central venous catheters when determining the probability of deep vein thrombosis.
Prevention of VTE in nonsurgical patients: Antithrombotic Therapy and Prevention of Thrombosis, 9th ed: American College of Chest Physicians Evidence-Based Clinical Practice Guidelines.
Yet, it is unknown whether peripherally inserted central catheters moderate an increase in venous thromboembolism through upper- or lower-extremity deep vein thrombosis or pulmonary embolism. We know of no studies that have assessed whether peripherally inserted central catheters are independently associated with an increase in the risk of subsequent lower-extremity deep vein thrombosis.
Therefore, we examined the association between peripherally inserted central catheter placement and subsequent venous thromboembolic events. We hypothesized that peripherally inserted central catheters would be associated with an increased risk of upper- and lower-extremity deep vein thrombosis and that this association would persist after adjustment for clinically important covariates.
Materials and Methods
Study Setting and Participants
The Michigan Hospital Medicine Safety consortium is a statewide quality improvement initiative that aims to prevent adverse events in hospitalized medical patients by creation of a data registry and sharing of best practices. The setting and design of this multicenter retrospective cohort study have been described.
Although participation is voluntary, each hospital receives payment for participating in the consortium and data collection.
Eligible cases included patients admitted to a medicine service for 2 days or longer. Patients meeting any of the following criteria were excluded: (1) aged less than 18 years; (2) pregnant; (3) underwent a surgical procedure during the admission; (4) directly admitted to an intensive care unit (ICU); (5) directly admitted for palliative care; (6) diagnosed with venous thromboembolism in the 6 months before admission; (7) admitted for presumed venous thromboembolism; (8) admitted under observation status; or (9) readmitted within 90 days of discharge from an admission included in this study.
Clinical data were collected at each hospital by dedicated, trained medical record abstractors. Patients discharged from each participating hospital were sampled on an 8-day rolling cycle to avert bias in selecting cases for review.
At each hospital, data on the first 18 eligible cases discharged during each cycle were collected. Follow-up data were obtained by both medical record review and direct telephone follow-up using a standardized script 90 days postdischarge. Each hospital undergoes annual in-person audits to ensure completeness and accuracy of data.
Covariates of Interest
Detailed demographics, medical history, comorbidities, physical examination findings, and laboratory and medication data using standardized definitions were collected by medical record review. Immobilization was defined if any of the following were documented in the medical record: bed rest ≥72 hours before hospitalization, immobilizing plaster cast, or paralysis. Similarly, clinical parameters such as current leg swelling were recorded on the basis of documentation in the medical chart. Deep vein thrombosis prophylaxis was defined as receipt of any of the following on day 1 or 2 of the index hospitalization: heparin 5000 units 2 times per day; heparin 5000 units 3 times per day; enoxaparin 40 mg/d; enoxaparin 30 mg/d (for creatinine clearance <30 mL/min); enoxaparin 30 mg 2 times per day; dalteparin 5000 units per day; or fondaparinux 2.5 mg/d.
Admissions to any hospital within 1 year before a patient entering the study were identified when documented in medical records. Only peripherally placed catheters that terminated at or close to the cavoatrial junction were categorized as peripherally inserted central catheters. Both peripherally inserted central catheters present on admission or inserted during the index hospitalization were included.
Ascertainment of Outcomes
All patients were followed for 90 days after the index hospitalization to identify the main outcomes of interest: symptomatic, image-confirmed, proximal upper-extremity deep vein thrombosis (involving the brachial, axillary, or subclavian veins), proximal lower-extremity deep vein thrombosis (iliac, femoral, popliteal veins), and pulmonary embolism. To be attributable to the peripherally inserted central catheter, events must have occurred after the date of peripherally inserted central catheter placement. Deep vein thrombosis was confirmed by Doppler ultrasound or venography, whereas pulmonary embolism was confirmed via computed tomography, ventilation perfusion scan, or pulmonary angiography. Venous thrombosis events that contributed to death or were the reason for transfer to the ICU were captured and included. Medical record review at 90 days was completed in 100% of eligible patients; telephone follow-up was completed in 58% of patients.
Statistical Analysis
Differences in patient characteristics by peripherally inserted central catheter status were assessed via chi-square tests. Cox proportional hazard models with Gamma-shared frailty by hospital (to account for clustering within the 48 hospitals) were fit to assess whether the risk-adjusted hazard of thrombosis outcomes differed by peripherally inserted central catheter status. The proportional hazards assumption by peripherally inserted central catheter status was assessed via the global test based on Schoenfeld residuals.
The primary outcome was time to venous thromboembolism from admission date of the index hospitalization. The primary exposure variable was the presence of a peripherally inserted central catheter. Three separate models were fit for time to each of the following outcomes: upper-extremity deep vein thrombosis, lower-extremity deep vein thrombosis, and pulmonary embolism. For each model, patients were right-censored at the first occurrence of the specific type of thrombotic event being modeled, transfer to the ICU or palliative care, or death. Patients were right-censored at 90 days after admission to the index hospitalization if they did not develop the thrombotic event being modeled, were not transferred to the ICU or palliative care, and did not die during the 90-day study period. If a patient experienced multiple types of thromboses, only the specific type of event being modeled was considered. All models were adjusted for the following covariates: hospitalization in the year preceding the index hospitalization, age, gender, cancer diagnosis within the year before hospitalization, prior venous thromboembolism, immobility, leg swelling, surgery within the month of the index hospitalization, chronic obstructive pulmonary disease, diabetes mellitus, congestive heart failure, cardiac arrhythmia, hypertension, inflammatory bowel disease, sepsis, and pneumonia.
Additional Cox proportional hazard models specified a priori were fit for sensitivity analyses. In addition to the covariates included in the main models, sensitivity analysis models also adjusted for receipt of pharmacologic prophylaxis on day 1 or 2 of the index hospitalization, which we have demonstrated is a suitable proxy for receipt during hospitalization.
All analyses were performed in Stata Version 13.0 (StataCorp LP, College Station, Tex).
Ethical and Regulatory Oversight
Because the purpose of the Hospital Medicine Safety Consortium is to measure and improve the quality of existing care practices, the project has received a “Not Regulated” status by the University of Michigan Medical School's Institutional Review Board.
Results
Data on 76,242 eligible patients were collected from January 2011 to March 2014 across 48 Michigan hospitals. The average age of patients was 66.7 years, and 41,812 (54.8%) were female. A total of 5613 patients (7.4%) had a history of cancer within the year preceding the index hospitalization, and 6159 patients (8.1%) had a history of venous thromboembolism. In addition, 36,690 patients (48.1%) had an inpatient stay in the year preceding the index hospitalization.
With respect to peripherally inserted central catheter use, 3790 patients (5.0%) were identified as having a peripherally inserted central catheter present at the time of hospital admission (n = 898) or placed during the hospital stay (n = 2892). Patients with peripherally inserted central catheters were more likely to have had prior inpatient stays, cancer diagnosis within the last year, prior venous thromboembolism, immobility, leg swelling during hospitalization, surgery within 30 days, diabetes, inflammatory bowel disease, sepsis, and pneumonia compared with those who did not receive this device (Table 1).
Table 1Select Patient Characteristics, Stratified by Peripherally Inserted Central Catheter Status (N = 76,242)
A total of 876 venous thromboembolism events occurred in 774 unique patients. Isolated, proximal upper-extremity deep vein thrombosis, lower-extremity deep vein thrombosis, and pulmonary embolism occurred in 182, 279, and 213 patients, respectively. Two patients experienced all 3 thrombotic events, 17 patients had both upper- and lower-extremity deep vein thrombosis, 74 patients developed lower-extremity thrombosis and pulmonary embolism, and 7 patients developed upper-extremity thrombosis and pulmonary embolism. Incidence rates of each venous thromboembolism outcome by peripherally inserted central catheter presence are shown in Table 2. Kaplan-Meier survival estimates for each type of venous thromboembolism outcome by peripherally inserted central catheter status are illustrated in Figure. Survivor functions differed by peripherally inserted central catheter status for each outcome (log-rank test for equality: proximal upper-extremity deep vein thrombosis, P < .001; proximal lower-extremity deep vein thrombosis, P = .0006; pulmonary embolism, P = .016).
Table 2Incidence of Venous Thromboembolism (per 10,000 Patient-days) Stratified by Peripherally Inserted Central Catheter Presence
Incidence rates for “Any VTE” are based on number of unique patients experiencing at least 1 type of VTE. A total of 100 patients experienced >1 type of VTE; as such, the total number of events and number of patient-days are not summative.
128/304,281 (4.21)
646/6,157,838 (1.05)
774/6,462,119 (1.20)
<.001
Results in each cell displayed as number of respective VTE events/number of patient-days (incidence rate per 10,000 patient-days).
DVT = deep vein thrombosis; PICC = peripherally inserted central catheter; VTE = venous thromboembolism.
∗ P values represent comparisons by PICC presence. The incidence rate is significantly greater among those with PICCs for each VTE outcome.
† Incidence rates for “Any VTE” are based on number of unique patients experiencing at least 1 type of VTE. A total of 100 patients experienced >1 type of VTE; as such, the total number of events and number of patient-days are not summative.
FigureKaplan-Meier curve of venous thromboembolism-free survival, stratified by peripherally inserted central catheter status. A, Proximal upper-extremity deep vein thrombosis. B, Proximal lower-extremity deep vein thrombosis. C, Pulmonary embolism. A, Event-free survival for proximal upper-extremity deep vein thrombosis, stratified by peripherally inserted central catheter status (solid line = no peripherally inserted central catheter; dashed line = peripherally inserted central catheter). B, Event-free survival for proximal lower-extremity deep vein thrombosis, stratified by peripherally inserted central catheter status (solid line = no peripherally inserted central catheter; dashed line = peripherally inserted central catheter). C, Event-free survival for pulmonary embolism, stratified by peripherally inserted central catheter status (solid line = no peripherally inserted central catheter; dashed line = peripherally inserted central catheter). DVT = deep vein thrombosis; PICC = peripherally inserted central catheter.
Among the 76,242 eligible patients, 1460 (1.91%) were transferred to the ICU for reasons other than venous thromboembolism or transitioned to palliative/comfort care. A total of 4728 patients (6.20%) died between admission to the index hospitalization and 90 days.
Proximal Upper-extremity Deep Vein Thrombosis
A total of 6,484,408 patient-days were examined for the outcome of proximal upper-extremity deep vein thrombosis, which occurred in 208 patients. A total of 1458 patients were censored because of transfer to the ICU or palliative care, and 4643 patients were censored because of death. The risk-adjusted hazard of upper-extremity deep vein thrombosis was 10-fold greater in patients with peripherally inserted central catheters compared with those without a peripherally inserted central catheter (hazard ratio [HR], 10.49; 95% confidence interval [CI], 7.79-14.11; P < .001). Prior hospitalization, cancer diagnosis in the year preceding the index hospitalization, and prior venous thromboembolism were associated with an increased hazard of proximal upper-extremity deep vein thrombosis (Table 3).
Proximal Lower-extremity Deep Vein Thrombosis
A total of 6,478,380 patient-days were examined for the outcome of proximal lower-extremity deep vein thrombosis, which occurred in 372 patients. A total of 1454 patients were censored because of transfer to the ICU or palliative care, and 4610 patients were censored because of death. The risk-adjusted hazard of lower-extremity deep vein thrombosis was greater in patients with peripherally inserted central catheters compared with those without peripherally inserted central catheters (HR, 1.48; 95% CI, 1.02-2.15; P = .038). Age ≥70 years, prior hospitalization, cancer diagnosis in the year before the index hospitalization, history of venous thromboembolism, immobility, leg swelling, and admitting diagnosis of pneumonia were associated with greater risk of lower-extremity deep vein thrombosis (Table 3).
Pulmonary Embolism
A total of 6,481,069 patient-days were examined for the outcome of pulmonary embolism, which occurred in 296 patients. In addition, 1456 patients were censored because of transfer to the ICU or palliative care, and 4630 patients were censored because of death. The risk-adjusted hazard of pulmonary embolism did not differ between patients with peripherally inserted central catheters compared with those without peripherally inserted central catheters (HR, 1.34; 95% CI, 0.86-2.06; P = .193). Prior hospitalization, cancer diagnosis in the year before the index hospitalization, and leg swelling during admission were associated with greater hazard of pulmonary embolism (Table 3).
Table 3Risk-adjusted Hazard Ratios Stratified by Venous Thromboembolism Type (N = 76,242)
Any VTE based on number of unique patients experiencing at least 1 type of VTE. A total of 100 patients experienced >1 type of VTE; as such, the total number of events and number of patient-days are not summative.
∗ PICC presence was defined as present on admission or inserted during the index hospitalization.
† Immobility defined as having any of the following: bed rest ≥72 hours before hospitalization, immobilizing plaster cast, or paralysis.
‡ Any VTE based on number of unique patients experiencing at least 1 type of VTE. A total of 100 patients experienced >1 type of VTE; as such, the total number of events and number of patient-days are not summative.
To examine whether results were robust to the use of pharmacologic deep vein thrombosis prophylaxis during hospitalization, we excluded 15,278 patients (20.0%) who were receiving systemic anticoagulation or had elevated international normalized ratio values on admission. Pharmacologic prophylaxis on day 1 or 2 of hospitalization was administered to 38,580 (63.3%) of the 60,694 eligible patients. No statistically significant associations between receipt of pharmacologic prophylaxis and any thrombotic outcomes were noted (Appendix, available online).
Discussion
This study of more than 70,000 hospitalized general medical patients in 48 Michigan hospitals reports that peripherally inserted central catheters are strongly associated with upper-extremity deep vein thrombosis. A modest association between peripherally inserted central catheters and subsequent lower-extremity thrombosis was observed. Predictors such as cancer, immobilization, and prior hospitalization were differentially associated with each of these thrombosis outcomes. Moreover, receipt of pharmacologic prophylaxis during hospitalization did not influence the risk of subsequent venous thromboembolism in patients with peripherally inserted central catheters. Taken together, these findings suggest that the thrombotic burden associated with peripherally inserted central catheters may not be restricted to the extremity where the device resides or easily attenuated after insertion. Rather, a paradigm that incorporates (1) risks, benefits, and alternatives to peripherally inserted central catheter use; (2) deep vein thrombosis risk assessment before peripherally inserted central catheter placement; and (3) lower threshold of testing for arm or leg thrombosis in patients with peripherally inserted central catheters seems necessary.
The association between peripherally inserted central catheters and upper-extremity deep vein thrombosis is well known.
Upper extremity venous thrombosis in patients with cancer with peripherally inserted central venous catheters: a retrospective analysis of risk factors.
A meta-analysis pooling available data showed that peripherally inserted central catheters are associated with a 2.5-fold greater risk of thrombosis than nontunneled central venous catheters.
Risk factors for catheter-related thrombosis (CRT) in cancer patients: a patient-level data (IPD) meta-analysis of clinical trials and prospective studies.
However, what is novel and noteworthy in this study is that peripherally inserted central catheter presence also was associated with lower-extremity deep vein thrombosis. Although we cannot explain the mechanism behind this association, 3 hypotheses may help rationalize this observation. First, peripherally inserted central catheters may serve as a potent trigger for systemic thrombosis, setting in motion a cascade that manifests in various territories. Supportively, biomarkers such as d-dimer and fragment 1+2 are known to both presage and increase with catheter-related thrombosis.
Whether these markers are by-products or actual mediators of thrombotic events related to catheters is unknown. Dedicated prospective studies examining the kinetics of these and related thrombosis markers in patients with peripherally inserted central catheters seem necessary.
Second, it is possible that the observed relationship between peripherally inserted central catheters and lower-extremity thrombosis relates to differential diagnostic ascertainment. For example, prior venous thrombosis is a well-recognized predictor of future thrombotic events.
Diagnosis of DVT: Antithrombotic Therapy and Prevention of Thrombosis, 9th ed: American College of Chest Physicians Evidence-Based Clinical Practice Guidelines.
The risk of recurrent venous thromboembolism after discontinuing anticoagulation in patients with acute proximal deep vein thrombosis or pulmonary embolism. A prospective cohort study in 1,626 patients.
Therefore, patients diagnosed with lower-extremity thrombosis may have been rendered susceptible to this outcome by first experiencing peripherally inserted central catheter-related thrombosis, an outcome that is often clinically silent.
Accordingly, patients labeled as experiencing leg deep vein thrombosis may have developed this event as a result of proximal upper-extremity thrombosis, leading to inaccurate attribution.
Because we did not screen for asymptomatic thrombotic events, the influence of clinically quiescent arm thrombosis on subsequent events cannot be determined from this study.
Finally, the association between peripherally inserted central catheter use and lower-extremity deep vein thrombosis may be explained by the fact that peripherally inserted central catheters are used often in high-risk patients, those inherently predisposed to thrombosis.
Are peripherally inserted central catheters associated with increased risk of adverse events in status 1B patients awaiting transplantation on continuous intravenous milrinone?.
Comparison of catheter-related large vein thrombosis in centrally inserted versus peripherally inserted central venous lines in the neurological intensive care unit.
Also, certain acute inflammatory illnesses (eg, pneumonia) increase the risk of lower-extremity deep vein thrombosis and often require peripherally inserted central catheter use.
However, because the association between peripherally inserted central catheters and lower-extremity deep vein thrombosis persists after adjusting for numerous known and unknown risk factors, confounding may not explain our findings. Nonetheless, the limited strength of the observed association, lack of convincing prior evidence, and presence of alternative explanations suggest that it would be premature to categorize the relationship between peripherally inserted central catheter use and lower-extremity deep vein thrombosis as causal.
Study Limitations
First, this is an observational study limited to general medical patients in non-ICU settings. Our findings are thus subject to confounding from unmeasured variables and limitations in external validity. Second, data regarding peripherally inserted central catheter characteristics and care practices were not collected; thus, whether or how such elements influence the risk of venous thromboembolism is not known. Third, we defined peripherally inserted central catheter exposure as patients who were admitted with peripherally inserted central catheters and patients who received such devices during hospitalization. Although these represent clinically disparate populations, we adjusted for comorbidities and hospitalization before the index stay to account for confounding from preexisting illness.
These limitations notwithstanding, our study has important strengths. First, this is the first and largest study to report associations between peripherally inserted central catheter use and 3 distinct venous thrombosis outcomes in hospitalized general medical patients. Our results are timely given the growing recognition of the nexus between peripherally inserted central catheters and hospital-acquired venous thromboembolism. Second, we observed that modifiable (eg, infections requiring antibiotics, such as pneumonia) and nonmodifiable patient characteristics (eg, prior venous thrombosis) were differentially associated with thrombotic events. Such observations may help inform and tailor the use of peripherally inserted central catheters in hospitalized patients. Third, our study helps inform clinical practices during hospitalization. Careful weighing of the risks and benefits of peripherally inserted central catheter use and consideration of alternative devices in those at high-risk of deep vein thrombosis seem essential. Of note, clinicians should not focus only on the extremity where a peripherally inserted central catheter resides, but the composite risk of venous thromboembolism among patients who receive a peripherally inserted central catheter.
Conclusions
We found that peripherally inserted central catheter insertion is associated with both upper- and lower-extremity deep vein thrombosis. Future studies that better define mechanisms for this association seem necessary. In the interim, mindful weighing of the thrombotic risks of peripherally inserted central catheters against their clinical benefits would be welcomed.
Appendix. Risk-Adjusted Hazard Ratios by Venous Thromboembolism Type: Sensitivity Analyses Examining Receipt of Pharmacologic Prophylaxis (N = 60,964)
Prevention of VTE in nonsurgical patients: Antithrombotic Therapy and Prevention of Thrombosis, 9th ed: American College of Chest Physicians Evidence-Based Clinical Practice Guidelines.
Upper extremity venous thrombosis in patients with cancer with peripherally inserted central venous catheters: a retrospective analysis of risk factors.
Risk factors for catheter-related thrombosis (CRT) in cancer patients: a patient-level data (IPD) meta-analysis of clinical trials and prospective studies.
Diagnosis of DVT: Antithrombotic Therapy and Prevention of Thrombosis, 9th ed: American College of Chest Physicians Evidence-Based Clinical Practice Guidelines.
The risk of recurrent venous thromboembolism after discontinuing anticoagulation in patients with acute proximal deep vein thrombosis or pulmonary embolism. A prospective cohort study in 1,626 patients.
Are peripherally inserted central catheters associated with increased risk of adverse events in status 1B patients awaiting transplantation on continuous intravenous milrinone?.
Comparison of catheter-related large vein thrombosis in centrally inserted versus peripherally inserted central venous lines in the neurological intensive care unit.
Funding: This study was funded by Blue Cross Blue Shield of Michigan. Other than research support, no compensation was received for this project. VC is supported by a career development award from the Agency for Healthcare Research and Quality (1K08HS022835-01). Blue Cross Blue Shield of Michigan supported data collection at each participating site and funded the data coordinating center but had no role in the study concept, interpretation of findings, or preparation, review, or final approval of the manuscript.
Conflict of Interest: SAF discloses consultancies for the Institute for Healthcare Improvement and the Society of Hospital Medicine; royalties from Wiley Publishing; honoraria for various talks at hospitals as a visiting professor; grants from the Centers for Disease Control Foundation, Blue Cross Blue Shield of Michigan, and Michigan Hospital Association; and expert witness testimony. SCW discloses serving as a panelist for the American College of Chest Physicians Clinical Practice Guideline: Antithrombotic Therapy for Venous Thromboembolic Disease (AT10). SJB discloses memberships in the Blue Care Network Statewide Clinical Quality Committee and the Blue Care Network Medical Leadership Council.
Authorship: All authors had access to the data and played a role in writing this manuscript.
00319-8&id=gr1.jpg){kind=link}