Management of epistaxis in patients with ventricular assist device: a retrospective review

Epistaxis Rates and Health Care Utilization in Patients With a Ventricular Assist Device

Objective

Identify baseline epistaxis rates and epistaxis-related health care utilization trends in the ventricular assist device (VAD) population.

Methods

Single center, retrospective cohort study consisting of chart review of adult VAD patients. Analysis of descriptive statistics was assessed using χ 2 tests, independent sample t tests, or Fisher's exact when expected counts were low. Logistic regression was used to assess associations between epistaxis and variables of interest.

Results

Two hundred ninety patients were included in the analysis. Ninety-eight (33.8%) patients developed epistaxis and 84 (29.0%) received medical attention. Patients with gastrointestinal (GI) bleeding had increased rates of epistaxis (42.4% vs 29.0%). Logistic regression analysis found GI bleeding to have an adjusted odds of developing epistaxis of 1.94 (95% confidence interval [CI]: 1.12-3.37) and kidney disease to have an adjusted odds of 1.83 (95% CI: 1.06, 3.13).

Discussion

VAD implantation improves survival and quality of life but also carries significant bleeding risks. At our institution, 29% of VAD patients received medical attention for epistaxis. GI bleeding and kidney disease were found to have increased adjusted odds of developing epistaxis. Fifty-nine percent of epistaxis events occurred while inpatient and 32.8% of events were seen in the emergency department.

Implications for Practice

VAD patients are an at-risk group that could potentially benefit from preventative nasal hydration regimen.

Device-related epistaxis risk: continuous-flow left ventricular assist device-supported patients

BACKGROUND

The aim of this study was to analyze the effect of device-dependent factors on epistaxis episodes comparing patients supported with a continuous-flow left ventricular assist device (CF-LVAD) to patients under the same antithrombotic therapy.

METHODS

Patients who underwent CF-LVAD between 2012 and 2018 were reviewed retrospectively from the institutionally adopted electronic database. Patients who underwent mitral valve replacement (MVR) surgery receiving the same anticoagulant and antiaggregant therapy were included as a control group. Demographics, epistaxis episodes, and nonepistaxis bleeding between the two groups were compared.

RESULTS

A total of 179 patients met the inclusion criteria (61 patients CF-LVAD group, 118 patients MVR group). The median (range) follow-up periods for the study (CF-LVAD) and control (MVR) groups were 370 (2819) and 545.70 (2356) days, respectively. There was a significant difference for frequency of bleeding episodes per month between CF-LVAD and MVR groups (p = 0.003 < 0.05). The most common site of bleeding was the anterior septum in both groups (90.9% for the CF-MVR group and 100% for the MVR group). While 14 patients (23%) had nonepistaxis bleeding in the CF-LVAD group, only two patients (1.7%) had nonepistaxis bleeding in the MVR group. There were significant differences in nonepistaxis bleeding rates between the CF-LVAD and MVR groups (χ²=19.79, p < 0.001).

CONCLUSION

Both epistaxis and nonepistaxis bleeding rates were higher in the CF-LVAD group than in the MVR group. This suggests that the use of CF-LVAD support could directly increase the risk of hemorrhagic complications.

LEVEL OF EVIDENCE

2A (Etiology/Harm).