Costs and Complications of Hospital Admissions for Inferior Vena Cava Filter Dysfunction

Abstract

Inferior vena cava filter (IVCF) dysfunction can result from penetration, fracture, and migration of the device necessitating retrieval. Endovascular and open removal of IVCF have been described in institutional series without comparison. This study examines national hospital admissions for IVCF dysfunction and compares the outcomes of open and endovascular retrieval. The National Inpatient Sample database (2016-2017) was reviewed for admissions with International Classification of Diseases, 10th edition (ICD-10) codes specific for IVCF dysfunction (Table I). All ICD-10 procedural codes were reviewed, and patients were divided based on open or endovascular IVCF retrieval. Patient characteristics, outcomes, and costs of hospitalization were compared between the two groups. There were 535 patients admitted with a diagnosis of IVCF dysfunction. Open IVCF removal was performed in 100 patients and endovascular removal in 90 patients. (Table II). Of those undergoing open surgery, 45 (45%) patients required median sternotomy and 55 (55%) required abdominal surgeries. Most patients were white females with a mean age of 54.4 years (range, 49.3-59.6 years) with a history of deep venous thrombosis (55.3%) or pulmonary embolism (31.6%). Most patients with IVCF dysfunction were treated in large (81.6%) or urban teaching (94.7%) hospitals situated most commonly in the South (42.1%) and Northeast (29.0%) with no difference in characteristics of the patients or the centers between the two groups. Patients undergoing open IVCF removal were more likely to undergo surgery on elective basis compared to endovascular IVCF removal (75.0% vs 11.1%; P < .001). There was a trend toward higher infectious complications and overall complications with endovascular removal but that did not reach statistical significance. Open removal was associated with a mortality of 5.0% compared to no inpatient mortality with endovascular removal (P = .33). The mean hospital length of stay was 6.7 days (range, 5.1-8.3 days) with no difference between the two groups. Open removal was associated with significantly higher hospital costs than endovascular removal ($34,276 vs $19,758; P = .05). Filter removal for patients with IVCF dysfunction is associated with significant morbidity, mortality and cost regardless of modality of retrieval. The introduction of specific ICD-10 codes for IVCF dysfunction allows researchers to study these relatively rare events. Developing effective tools for outpatient retrieval of dysfunctional IVCF could decrease related hospitalization and have potential savings on the health care system.TableInternational Classification of Diseases, 10th edition (ICD-10) diagnosis codes for inferior vena cava dysfunctionICD-10 codeDescriptionT82.515ABreakdown (mechanical) of umbrella device, initial encounterT82525ADisplacement of umbrella device, initial encounterT82535ALeakage of umbrella device, initial encounter Open table in a new tab Table IICharacteristics of patients with inferior vena cava filter IIVCF) dysfunction with removal by surgery typeEndovascularOpenP valuen = 90n = 100Demographics Age57.4 [49.6-65.2]51.7 [45.0-58.5].28 Female65 (72.2%)65 (65.0%).66Race.24 White45 (50.0%)75 (75.0%) Black20 (22.2%)5 (5.0%) Hispanic5 (5.6%)10 (10.0%)Other90 (22.2%)10 (10.0%)Primary payer.09 Medicare65 (72.2%)35 (35.0%) Medicaid10 (11.1%)15 (15.0%) Private10 (11.1%)45 (45.0%) Other5 (5.6%)5 (5.0%)Median household income.28 $1-$42,99940 (44.4%)20 (21.1%) $43,000-$53,99910 (11.1%)35 (36.8%) $54,000-$70,99910 (11.1%)10 (10.5%) $71,000 or more30 (33.3%)30 (31.6%)Hospital region.64 Northeast30 (33.3%)25 (25.0%) Midwest15 (16.7%)15 (15.0%) South40 (44.4%)40 (40.0%) West5 (5.6%)20 (20.0%)Location/teaching status of hospital.94 Rural0 (0.0%)0 (0.0%) Urban nonteaching5 (5.6%)5 (5.0%) Urban teaching85 (94.4%)95 (95.0%)Hospital bed size.45 Small0 (0.0%)5 (5.0%) Medium10 (11.1%)20 (20.0%) Large80 (88.9%)75 (75.0%)Comorbidities Hypertension35 (38.9%)30 (30.0%).57 Coronary artery disease10 (11.1%)5 (5.0%).49 Congestive heart failure5 (5.6%)5 (5.0%).94 Diabetes without complication0 (0.0%)10 (10.0%).16 Diabetes with complications5 (5.6%)10 (10.0%).61 Chronic kidney disease15 (16.7%)10 (10.0%).54 Chronic obstructive pulmonary disorder15 (16.7%)15 (15.0%).89 Deep venous thrombosis50 (55.6%)55 (55.0%).97 Pulmonary embolism25 (27.8%)35 (35.0%).64 Charlson comorbidity index1.5 [0.8-2.2]1.1 [0.4-1.9].51 Elective10 (11.1%)75 (75.0%)<.001Outcomes Neurologic complication001.00 Respiratory complication5 (5.6%)10 (10.0%).61 Cardiac complication001.00 Thrombotic/embolic complication001.00 Bleeding complication001.00 Gastrointestinal complication5 (5.6%)5 (5.0%).94 Renal complication15 (16.7%)5 (5.0%).24 Infectious complication15 (16.7%)0.06 Shock0 (0.0%)5 (5.0%).33 Any complication35 (38.9%)15 (15.0%).10 Length of stay6.9 [4.1-9.8]6.5 [4.7-8.3].79 Total cost ($)19,758 [12,331-27,185]34,276 [21,606-46,945].05 In-hospital mortality05 (5.0%).33Values are median [interquartile range] unless otherwise noted. Open table in a new tab