Complications After Lumbar Spine Epidural Steroid Injections: Incidence and Risk Factors

Abstract

Study Design

Retrospective Study.

Objectives

Lumbar epidural steroid injections (LESIs) are frequently used to manage symptoms and pain stemming from degenerative conditions of the lumbar spine. This study aimed to determine the incidence and risk factors of complications after LESI.

Method

The Merative MarketScan Commercial and Medical Supplemental databases were queried for LESIs in 18+ patients. LESIs were distinguished based on the approach: interlaminar or transforaminal. Complications were categorized into overall, procedural, and medical. Generalized estimating equations logistic regression with repeated measures, clustered by patient identifier, were used to identify factors associated with complications at 7 and 30 days after the LESI.

Results

The study cohort consisted of 362,976 patients who underwent a total of 722,366 LESIs from 2014 to 2021. Complication rates after LESIs were 2.5% within 7 days and 8.8% within 30 days. 45.0% of LESIs utilized the interlaminar approach, while the transforaminal approach was used in 54.0% of LESIs. The patient factors independently associated with the greatest odds of complications included older age [85+ vs 18-34, 7 day OR: 1.32 (1.20, 1.45); P < 0.001], heart failure [7 day OR: 1.95 (1.84, 2.06); P < 0.001], renal dysfunction [7 day OR: 1.51 (1.41, 1.63); P < 0.001], neurological deficits [7 day OR: 1.37 (1.31, 1.44); P < 0.001], and anticoagulant prescriptions [7 day OR: 3.28 (3.10, 3.48); P < 0.001].

Conclusions

LESIs were associated with a rare but non-negligible risk of complications, which continued to occur until 30-days post injection. Several patient factors were associated with the risk of post-procedure complications.

Keywords

lumbar spine epidural steroid injections transforaminal interlaminar risk factors incidence complications

Introduction

Lumbar epidural steroid injections (LESIs) are one of the most frequently performed spinal procedures in the United States (US). An estimated 2.2 million ESIs are performed annually in the Medicare population.^1,2,3 LESIs are most commonly utilized in the management of painful lower back conditions, including spinal stenosis, disc herniations, and degenerative disc disease.⁴ Through delivering corticosteroids directly into the epidural space, LESIs aim to reduce inflammation of the nerve roots and surrounding tissues, which in turn alleviates pain and improves function in patients with lumbar spinal disease.^1,5,6 LESIs often serve as a temporary measure to reduce the need for surgery.⁷

The procedure can be performed via two primary techniques – interlaminar or transforaminal – each with particular anatomic targets and associated risks. Many reported complications from both LESI techniques are often minor and self-limiting, including increased injection site or back pain, vasovagal reaction, vascular penetration, transient nerve root pain, or entry into a site other than the epidural space (eg, facet joint, intervertebral disc).^8,9,10,11 However, while quite rare, more serious complications from these procedures have arisen,8⁸ causing the FDA to issue a letter of warning in 2014 stating that “the injection of corticosteroids into the epidural space may result in rare but serious adverse events, including loss of vision, stroke, paralysis, and death”.^12,13 Cases of these adverse complications occurring in practice have been well-documented in the literature.^{14,15,16,17,18,19} Specifically, while central nervous system complications like cerebral infarction and vision loss are more common with cervical ESIs, LESIs pose other potential significant complications including severe infections,^10,20 epidural hematomas,^21,22,23 and neurological injury.²⁴

Although several case reports and review articles have been published describing the potentially catastrophic effects of LESIs, the incidence rates of the specific complications are not well described. Additionally, there is a lack of literature determining the potential risk factors associated with complications from LESI procedures. The purpose of this study was to utilize a large dataset of administrative claims to characterize and quantify the incidence of complications after LESIs as well as to determine potential risk factors for developing those complications.

Methods

Design and Cohort

In this retrospective study, the Merative MarketScan Commercial Claims and Encounters and Medicare Supplemental database were queried. These databases are one of the largest collections of United States healthcare claims, containing de-identified data from more than 215 million patients derived from employers, hospitals, electronic medical records providers, and healthcare administrators.²⁵ This study was not considered human subjects research since the data used in this study were de-identified. The cohort consisted of LESI procedures from January 1, 2014 through September 30, 2021 in patients aged 18 and older. LESIs needed to be associated with a diagnosis of lumbar radiculopathy, lumbar spinal stenosis, lumbar herniated disc, degenerative disc disease, lumbar spondylosis, myelopathy of the lumbar spine, sciatica, post-laminectomy syndrome, and other non-specific lumbar spine conditions. Procedures were identified using Current Procedural Terminology codes, and diagnoses were identified using the International Classification of Diseases-9th revision and 10th revision (specific codes in Supplemental Table S1). For the LESI to be eligible for inclusion, the patient needed to be continuously enrolled in the claims database for at least 12 months prior to and 30 days after the LESI and to be enrolled in an insurance plan that reported outpatient pharmacy data to the claims database. Exclusion criteria were a lumbar spine surgery within the last 12 months before the LESI or having a diagnosis of a neoplasm, lumbar spinal trauma or fracture, infection, pregnancy, or inflammatory spondyloarthropathy (specific codes in Supplemental Table S1).

Variables of Interest

Patient demographic factors recorded were age, sex, United States (US) geographic region (Northeast, Midwest, South, and West), rural or urban location, and spinal diagnosis (of the inclusion diagnoses). Characteristics of the LESI included the year of injection, the injection approach (interlaminar, transforaminal, or both), whether it was a single or multilevel injection, the total number of LESI injections over the study period, and the timing between injections. Individual patient comorbidities were collected. The comorbidities were determined from the Elixhauser comorbidity index as well as studies by Wadhwa et al and Ratliff et al.^26,27,28 The comorbidities and definitions are listed in Supplemental Table S1. Patient prescription data for any medication at the time of the LESI was also gathered and reported, as well as specifically for prescribed anticoagulants, antiplatelet drugs, and non-steroidal anti-inflammatory drugs (NSAIDs) (see definitions in Supplemental Table S1). Active prescriptions at the time of a LESI were identified and tabulated based on the prescription fill date and days supplied.

The primary outcomes of this study were the overall complications at 7 days and 30 days after a LESI procedure, encompassing procedural complications, medical complications, hospital admission, or emergency department visit. Individual procedural and medical complications and their definitions are specified in Supplementa Table S1, with most previously defined by Harsha et al.²⁶ All-cause hospital admissions after injection and emergency department visits within 7 and 30 days were also recorded. If a patient had multiple LESIs in a short time frame, a complication was only attributed to the most recent LESI to avoid double-counting complications.

Statistical Analysis

Patient demographic and clinical factors, injection characteristics, and outcomes are reported as frequencies and percentages of the total number of LESIs. Chi-squared and Fisher’s exact tests were used to evaluate univariate differences in complication rates between interlaminar and transforaminal approaches of LESIs. Generalized estimating equations logistic regression with repeated measures, clustered by patient identifier, were employed to identify patient factors and injection characteristics that were associated with overall, medical, and procedural complications at 7 and 30 days. Each outcome and time point was modeled separately. Fixed effects included in the model were patient demographics, characteristics of the LESI, prescribed medications, and patient comorbidities. Results from the multivariable models were reported as odds ratios (ORs) and 95% confidence intervals (CIs). Modeling for the individual procedural outcomes of neurological injury and epidural hematoma was performed similarly. However, due to smaller event occurrences, epidural hematomas were only modeled at 30 days. All statistical tests were performed with SAS version 9.4 (Cary, NC, USA).

Results

Summary of Included LESIs

The final cohort included 722,366 LESI procedures in a total of 362,976 unique patients (Figure 1). Among all LESIs, the mean age of patients was 55.6 (SD = 14.1) years, and 53.2% were performed in females (Table 1). Almost half of LESIs occurred in patients living in the South (45.6%), and most (84.8%) occurred in those living in urban geographic areas. The most common diagnoses at the time of injection were lumbar radiculopathy (71.8%), lumbar disc herniation (26.5%), lumbar stenosis (19.9%), and lumbar degenerative disc disease (19.8%). The transforaminal approach was used in 54.0% of LESIs, interlaminar in 45.0%, and both approaches in 1.0%. Single-level injections were performed in 74.5% of LESIs, while 25.5% of LESIs were injected into multiple levels of the lumbar spine. For about half of LESIs (55.1%), there was no prior LESI within 12 months. For those with multiple LESIs, the median duration between injections was 57 days (IQR 23-140). In the cohort, 84.3% of LESIs were performed in those on some prescription medication. Specifically, 1.9% were on anticoagulants, 2.2% were on antiplatelet medications, and 20.7% were on NSAIDs at the time of their LESI (Table 1).

Figure 1.

A flow diagram visualizing the selection of lumbar epidural steroid injections for inclusion in this study

Table 1.

Characteristics of Study Cohort (n = 722,366)

Characteristics	n (%)
Age; mean (SD)	55.6 (14.1)
Sex
Female	384,058 (53.2)
Male	338,308 (46.8)
Region
Northeast	112,013 (15.5)
Midwest	195,861 (27.1)
South	329,576 (45.6)
West	84,916 (11.8)
Rural	109,624 (15.2)
Diagnoses of the lumbar spine
Radiculopathy	518,794 (71.8)
Stenosis	143,711 (19.9)
Herniated disc	191,269 (26.5)
Degenerative disc disease	143,208 (19.8)
Spondylosis	81,802 (11.3)
Myelopathy	26,622 (3.7)
Sciatica	34,707 (4.8)
Post-laminectomy syndrome	27,265 (3.8)
Other lumbar spine disorder	53,453 (7.4)
LESI details
Approach
Both	7302 (1.0)
Interlaminar	324,718 (45.0)
Transforaminal	390,346 (54.0)
Multiple levels	184,024 (25.5)
Injection year
2014	171,556 (23.7)
2015	139,725 (19.3)
2016	109,611 (15.2)
2017	78,145 (10.8)
2018	64,826 (9.0)
2019	64,850 (9.0)
2020	52,482 (7.3)
2021	41,171 (5.7)
Prior LESI in past year
0	398,260 (55.1)
1	191,077 (26.5)
2	83,298 (11.5)
3+	49,731 (6.9)
Days between injections; median (IQR)	57.0 (46.0-63.0)
Comorbidities
Depression	109,082 (15.1)
Anxiety	127,000 (17.6)
Other psychiatric conditions	64,827 (9.0)
Chronic obstructive pulmonary disease	35,523 (4.9)
Heart failure	18,522 (2.6)
Hypertension	363,266 (50.3)
Previous myocardial infarction	7788 (1.1)
Coronary artery disease	63,450 (8.8)
Peripheral artery disease	30,693 (4.2)
Anemia	68,239 (9.4)
Coagulation disorders	5286 (0.7)
Thrombocytopenia	6535 (0.9)
Other cardiac conditions	54,909 (7.6)
Other pulmonary conditions	92,754 (12.8)
Diabetes	138,132 (19.1)
Renal dysfunction	11,128 (1.5)
Obesity	122,995 (17.0)
Alcohol use	9380 (1.3)
Tobacco use	64,220 (8.9)
Other drug use	30,660 (4.2)
Neurological disorders	48,149 (6.7)
Hyperlipidemia	314,809 (43.6)
Gastroesophageal reflux disease	120,443 (16.7)
Chronic pain	177,793 (24.6)
Immunodeficiency	19,705 (2.7)
Inflammatory disease	38,094 (5.3)
Steroid use	3882 (0.5)
History of venous thromboembolism	1910 (0.3)
Medications
Any medication	609,079 (84.3)
Anticoagulants	13,943 (1.9)
Antiplatelets	15,963 (2.2)
Non-steroidal anti-inflammatory drugs	149,695 (20.7)

LESI, lumbar epidural steroid injection; SD, standard deviation; IQR, interquartile range.

Complication Rates

The rate of all complications was 2.5% and 8.8% at 7 and 30 days, respectively (Table 2). Procedural complications occurred at 0.3%, 0.9%, and 2.1%, respectively, with a total of 23,266 recorded during the study period. Neurological injury, including parasthesias, was the most common procedural complication, occurring in 0.2% and 0.8% of LESIs, respectively. Epidural hematomas, infections, and durotomies occurred in less than 0.1% of LESIs for both timepoints. Medical complications were more common, with a total of 101,221 recorded medical complications across the study period. Medical complications occurred in 1.3% and 4.2% of LESIs at 7 and 30 days, respectively. Dysrhythmias were the most common medical complication, with rates of 0.6% and 1.9%, respectively, followed by urinary tract infection (UTI) (0.3% and 1.3%, respectively). Heart failure occurred at a rate of 0.2% and 0.6% of LESIs, respectively, for both timepoints. All other medical complications occurred in less than 1% of LESI cases. Emergency department visits occurred in 1.1% and 4.2% of LESIs, respectively, and all-cause admission was recorded at 0.3% and 1.3%, respectively. (Table 2).

Table 2.

Incidence of Complications, Hospital Readmission, and Emergency Department Visits at 7 and 30 days After Lumbar Epidural Steroid Injection Procedures

Complication	7-day	30-day
Complication	n (%)	n (%)
Overall complications	18,324 (2.5)	63,501 (8.8)
Total procedural complications	1809 (0.3)	6464 (0.9)
Neurological injury	1710 (0.2)	5910 (0.8)
Air embolism	<11	<11
Epidural hematoma	31 (<0.1)	142 (<0.1)
Infection	29 (<0.1)	183 (<0.1)
Dural tear	23 (<0.1)	188 (<0.1)
Foreign body	<11	<11
Other procedural complication	46 (<0.1)	255 (0.0)
Total medical complications	9090 (1.3)	30,654 (4.2)
Dysrhythmia	4127 (0.6)	13,901 (1.9)
Pneumonia	482 (0.1)	1964 (0.3)
Renal failure	718 (0.1)	2142 (0.3)
Myocardial infarction	241 (<0.1)	799 (0.1)
Pulmonary embolism	262 (<0.1)	960 (0.1)
Heart failure	1175 (0.2)	4092 (0.6)
Deep vein thrombosis	589 (0.1)	2095 (0.3)
Urinary tract infection	2328 (0.3)	9034 (1.3)
Delirium	125 (<0.1)	377 (0.1)
Readmission	1972 (0.3)	9711 (1.3)
Emergency department visit	8106 (1.1)	30,266 (4.2)

Exact cell counts <11 are suppressed to comply with the MarketScan data use agreement.

When stratified by injection approach, overall complications occurred more frequently for interlaminar vs the transforaminal approach at 7 days (2.7% vs 2.4%, P < 0.001) and 30 days (9.2% vs 8.5%, P < 0.001) (Table 3). Procedural complications 7 days and 30 days post-LESI were more frequent when the interlaminar approach was utilized compared to a transforaminal approach (7 days: 0.3% vs 0.2%, P < 0.001; 30 days: 0.9% vs 0.9%, P < 0.001). The interlaminar approach was also associated with higher frequency of medical complications at 7 days (1.4% vs 1.2%, P < 0.001) and 30 days (4.6% vs 4.0%, P < 0.001). Frequencies of specific procedural and medical complications stratified by approach are presented in Table 3.

Table 3.

Incidence of Complications at 7 and 30 days After Lumbar Epidural Steroid Injection Stratified by Interlaminar Versus Transforaminal Approach

Complication	7- day			30- day
	IL	TF	P-value	IL	TF	P-value
	n (%)	n (%)	P-value	n (%)	n (%)	P-value
Any complication	8728 (2.7)	9423 (2.4)	<0.001	29,733 (9.2)	33,140 (8.5)	<0.001
Total procedural complications	940 (0.3)	852 (0.2)	<0.001	3053 (0.9)	3363 (0.9)	<0.001
Neurological injury	891 (0.3)	802 (0.2)	<0.001	2797 (0.9)	3068 (0.8)	<0.001
Air embolism	<11	<11	0.12	<11	<11	0.03
Epidural hematoma	23 (<0.1)	<11	0.001	>62 (<0.1)	69 (<0.1)	0.18
Infection	12 (<0.1)	17 (<0.1)	0.66	>78 (<0.1)	94 (<0.1)	0.47
Dural tear	14 (<0.1)	<11	0.086	>61 (<0.1)	116 (<0.1)	0.03
Foreign body	<11	<11	0.79	<11	<11	0.90
Other procedural Complication	21 (<0.1)	25 (<0.1)	0.97	>105 (<0.1)	139 (<0.1)	0.91
Total medical complications	4444 (1.4)	4563 (1.2)	<0.001	14,861 (4.6)	15,507 (4.0)	<0.001
Dysrhythmia	2042 (0.6)	2049 (0.5)	<0.001	6762 (2.1)	7018 (1.8)	<0.001
Pneumonia	>210 (0.1)	261 (0.1)	0.93	940 (0.3)	1013 (0.3)	0.02
Renal failure	>361 (0.1)	346 (0.1)	0.002	1076 (0.3)	1044 (0.3)	<0.001
Myocardial infarction	>111 (<0.1)	119 (<0.1)	0.14	>371 (0.1)	417 (0.1)	0.29
Pulmonary embolism	>130 (<0.1)	121 (<0.1)	0.009	>505 (0.2)	444 (0.1)	<0.001
Heart failure	>604 (0.2)	560 (0.1)	<0.001	2102 (0.6)	1955 (0.5)	<0.001
Deep vein thrombosis	>295 (0.1)	283 (0.1)	0.004	1037 (0.3)	1040 (0.3)	<0.001
Urinary tract infection	1100 (0.3)	1206 (0.3)	0.03	4352 (1.3)	4580 (1.2)	<0.001
Delirium	>53 (<0.1)	61 (<0.1)	0.27	>192 (0.1)	174 (0.0)	0.003
Readmission	951 (0.3)	1002 (0.3)	0.004	4492 (1.4)	5115 (1.3)	0.008
Emergency department visit	3745 (1.2)	4282 (1.1)	0.02	14,004 (4.3)	15,961 (4.1)	<0.001

Injections with both interlaminar and transforaminal approaches are not included in this table given low frequency in the cohort (1.0%). Cell counts <11 are suppressed to comply with the MarketScan data use agreement. Similarly, cell counts indicated as > x are coarsened secondary to small cell counts for the “both” approach group that could be calculated from a combination of the data in this table and Table 2.

Factors Associated with Overall Complications

In the multivariable model, many patient demographics had associations with increased odds of overall complications. Female sex was associated with an increased odds of complications at 7, and 30 days [7 day OR: 1.13 (1.09, 1.17); P < 0.001] (Table 4). Age was associated with increased odds of a complication, with patients aged 75 years and above having the highest odds (P < 0.001 for all timepoints). Younger patients had significantly lower odds of sustaining complications, especially patients between ages 45 and 64. The patient comorbidities with the largest association for complications at both timepoints included a diagnosis of congestive heart failure [7 day OR: 1.95 (1.84, 2.06); P < 0.001], other cardiac conditions [7 day OR: 1.57 (1.51, 1.64); P < 0.001], renal dysfunction [7 day OR: 1.51 (1.41, 1.63); P < 0.001], coagulation disorders [7 day OR: 1.37 (1.24, 1.52); P < 0.001], and neurological disorders [7 day OR: 1.37 (1.31, 1.44); P < 0.001]. A history of venous thromboembolism was associated with increased odds of complications at 30 days [1.43 (1.35, 1.51); P < 0.001]. Other associated comorbidities include depression, anxiety, hypertension, diabetes, tobacco use, psychiatric conditions, and alcohol use. Anticoagulant prescription was the pharmacologic factor most strongly associated with complications at 7 [3.28 (3.10, 3.48); P < 0.001] and 30 days [3.70 (3.55, 3.85); P < 0.001] post-LESI (Table 4).

Table 4.

Patient Demographics, Clinical Factors, and Injection Characteristics Associated With Any Complication at 7 and 30 Days

Characteristic	7-day		30-day
Characteristic	OR (95% CI)	P-value	OR (95% CI)	P-value
Age (Ref: 18-34)
35-44	0.84 (0.78, 0.91)	<0.001	0.86 (0.82, 0.89)	<0.001
45-54	0.69 (0.64, 0.74)	<0.001	0.77 (0.74, 0.80)	<0.001
55-64	0.72 (0.67, 0.77)	<0.001	0.77 (0.74, 0.80)	<0.001
65-74	0.85 (0.79, 0.93)	<0.001	0.92 (0.88, 0.96)	<0.001
75-84	1.06 (0.98, 1.15)	0.14	1.16 (1.10, 1.21)	<0.001
85+	1.32 (1.20, 1.45)	<0.001	1.43 (1.35, 1.51)	<0.001
Female sex (Ref: Male)	1.13 (1.09, 1.17)	<0.001	1.20 (1.18, 1.23)	<0.001
Region (Ref: Northeast)
Midwest	0.97 (0.93, 1.02)	0.26	0.88 (0.86, 0.90)	<0.001
South	0.96 (0.92, 1.01)	0.09	0.91 (0.89, 0.93)	<0.001
West	1.02 (0.96, 1.08)	0.47	0.92 (0.89, 0.95)	<0.001
Rural	0.98 (0.94, 1.02)	0.26	0.98 (0.96, 1.01)	0.20
Diagnoses of the lumbar spine
Radiculopathy	1.03 (0.99, 1.07)	0.14	1.00 (0.98, 1.02)	0.93
Stenosis	0.99 (0.96, 1.03)	0.79	1.01 (0.99, 1.04)	0.24
Herniated disc	1.01 (0.98, 1.05)	0.49	0.99 (0.97, 1.01)	0.45
Degenerative disc disease	1.01 (0.97, 1.05)	0.77	0.99 (0.97, 1.01)	0.50
Spondylosis	1.04 (0.99, 1.09)	0.15	1.01 (0.98, 1.04)	0.43
Myelopathy	0.96 (0.87, 1.05)	0.36	0.95 (0.90, 1.00)	0.06
Sciatica	1.08 (1.01, 1.16)	0.04	0.98 (0.94, 1.02)	0.36
Post-laminectomy syndrome	0.99 (0.92, 1.07)	0.82	1.03 (0.98, 1.07)	0.21
Other lumbar spine disorder	1.02 (0.96, 1.08)	0.46	1.03 (0.99, 1.06)	0.10
LESI details
Approach (Ref: Interlaminar)
Both	0.93 (0.80, 1.09)	0.37	0.99 (0.91, 1.08)	0.81
Transforaminal	0.97 (0.94, 1.00)	0.06	1.00 (0.98, 1.01)	0.63
Injection year (Ref: 2014)
2015	1.00 (0.96, 1.05)	0.91	0.98 (0.96, 1.01)	0.20
2016	0.96 (0.91, 1.01)	0.09	0.94 (0.92, 0.97)	<0.001
2017	0.87 (0.82, 0.92)	<0.001	0.91 (0.88, 0.94)	<0.001
2018	0.89 (0.84, 0.95)	<0.001	0.93 (0.89, 0.96)	<0.001
2019	0.95 (0.89, 1.01)	0.08	0.93 (0.90, 0.96)	<0.001
2020	0.84 (0.78, 0.90)	<0.001	0.87 (0.83, 0.90)	<0.001
2021	0.93 (0.86, 0.99)	0.03	0.94 (0.90, 0.98)	0.002
Prior LESI in past year	0.96 (0.95, 0.98)	<0.001	1.00 (0.99, 1.01)	0.76
Comorbidities
Depression	1.10 (1.05, 1.14)	<0.001	1.14 (1.11, 1.16)	<0.001
Anxiety	1.16 (1.12, 1.21)	<0.001	1.19 (1.16, 1.22)	<0.001
Other psychiatric conditions	1.25 (1.19, 1.31)	<0.001	1.24 (1.21, 1.27)	<0.001
COPD	1.05 (0.99, 1.11)	0.10	1.11 (1.07, 1.15)	<0.001
HF	1.95 (1.84, 2.06)	<0.001	2.17 (2.09, 2.25)	<0.001
HTN	1.29 (1.24, 1.34)	<0.001	1.23 (1.21, 1.26)	<0.001
Previous MI	1.05 (0.95, 1.15)	0.36	1.14 (1.07, 1.21)	<0.001
CAD	1.27 (1.21, 1.33)	<0.001	1.30 (1.26, 1.33)	<0.001
PAD	1.16 (1.10, 1.23)	<0.001	1.16 (1.12, 1.20)	<0.001
Anemia	1.30 (1.25, 1.35)	<0.001	1.21 (1.18, 1.24)	<0.001
Coagulation disorders	1.37 (1.24, 1.52)	<0.001	1.29 (1.20, 1.39)	<0.001
Thrombocytopenia	1.19 (1.07, 1.33)	0.002	1.16 (1.08, 1.25)	<0.001
Other cardiac conditions	1.57 (1.51, 1.64)	<0.001	1.52 (1.48, 1.56)	<0.001
Other pulmonary conditions	1.24 (1.19, 1.29)	<0.001	1.22 (1.19, 1.24)	<0.001
Diabetes	1.20 (1.16, 1.25)	<0.001	1.16 (1.13, 1.18)	<0.001
Renal dysfunction	1.51 (1.41, 1.63)	<0.001	1.48 (1.41, 1.56)	<0.001
Obesity	1.13 (1.08, 1.17)	<0.001	1.12 (1.10, 1.15)	<0.001
Alcohol use	1.12 (1.00, 1.25)	0.05	1.17 (1.14, 1.20)	<0.001
Tobacco use	1.21 (1.15, 1.27)	<0.001	1.17 (1.14, 1.20)	<0.001
Other drug use	1.20 (1.12, 1.28)	<0.001	1.19 (1.15, 1.24)	<0.001
Neurological disorders	1.37 (1.31, 1.44)	<0.001	1.36 (1.33, 1.40)	<0.001
Hyperlipidemia	0.97 (0.94, 1.00)	0.09	1.00 (0.98, 1.02)	0.68
GERD	1.14 (1.10, 1.18)	<0.001	1.15 (1.13, 1.18)	<0.001
Chronic pain	1.15 (1.11, 1.19)	<0.001	1.17 (1.15, 1.20)	<0.001
Immunodeficiency	1.27 (1.19, 1.37)	<0.001	1.27 (1.22, 1.33)	<0.001
Inflammatory disease	1.05 (0.99, 1.12)	0.11	1.08 (1.05, 1.12)	<0.001
Steroid use	1.20 (1.03, 1.40)	0.02	1.22 (1.11, 1.35)	<0.001
History of VTE	1.18 (0.99, 1.39)	0.06	1.43 (1.28, 1.60)	<0.001
Medications
Anticoagulants	3.28 (3.10, 3.48)	<0.001	3.70 (3.55, 3.85)	<0.001
Antiplatelets	1.12 (1.03, 1.21)	0.005	1.14 (1.09, 1.19)	<0.001
NSAIDs	0.98 (0.94, 1.02)	0.30	0.98 (0.96, 1.00)	0.04

OR, odds ratio; CI, confidence interval; LESI, lumbar epidural steroid injection; Ref, Reference; COPD, chronic obstructive pulmonary disease; HF, heart failure; HTN, hypertension; MI, myocardial infarction; CAD, coronary artery disease; PAD, peripheral artery disease; GERD, gastroesophageal reflux disease; VTE, venous thromboembolism; NSAID, non-steroidal anti-inflammatory drug.

For variables where a reference category is not specified, the comparator is yes vs no.

Factors Associated with Procedural Complications

Advanced age was determined to be one of the highest risk factors for procedural complications after a LESI procedure. Patients over the age of 85 years had the greatest odds for 30-day [3.28 (2.71, 3.97); P < 0.001] procedural complications vs ages 18-34 years (Table 5). Antiplatelet prescription was associated with increased odds for procedural complications at both timepoints post-LESI [7 day OR: 2.47 (2.10, 2.90); P < 0.001]. Residence in a rural area was associated with decreased odds for procedural complications for both timepoints [7 day OR: 0.86 (0.75, 0.99); P = 0.03]. After multivariate adjustment, the transforaminal approach was associated with a lower risk of procedural complications at 7 days [0.82 (0.74, 0.90); P < 0.001] compared to the interlaminar approach. The patient comorbidity with the largest associations was neurological disorders at 7 [2.67 (2.38, 3.00); P < 0.001] and 30 days [2.29 (2.15, 2.44); P < 0.001]. Other patient comorbidities associated with greater odds for procedural complications included peripheral artery disease [7 day OR: 1.64 (1.43, 1.89); P < 0.001], coagulation disorders [7 day OR: 1.62 (1.21, 2.18); P = 0.001], hypertension [7 day OR: 1.45 (1.28, 1.64); P < 0.001], hyperlipidemia [7 day OR: 1.33 (1.19, 1.48); P < 0.001], diabetes [7 day OR: 1.26 (1.13, 1.40); P < 0.001], and psychiatric conditions [7 day OR: 1.39 (1.20, 1.61); P < 0.001]. A previous myocardial infarction was associated with a 0.62 (0.45, 0.85) fold decrease in odds for procedural complications 7 days after an LESI procedure (P = 0.003). The odds for procedural complications were also elevated if a patient was prescribed anticoagulant medications [7 day OR: 1.77 (1.47, 2.14); P < 0.001] (Table 5).

Table 5.

Patient Demographics, Clinical Factors, and Injection Characteristics Associated With Procedural Complications at 7 and 30 Days

	7-day		30-day
Characteristic	OR (95% CI)	P-value	OR (95% CI)	P-value
Age (Ref: 18-34)
35-44	1.14 (0.82, 1.59)	0.44	1.00 (0.83, 1.21)	0.97
45-54	1.12 (0.82, 1.53)	0.47	1.21 (1.02, 1.43)	0.03
55-64	1.27 (0.94, 1.73)	0.12	1.46 (1.24, 1.72)	<0.001
65-74	1.99 (1.44, 2.74)	<0.001	2.39 (2.01, 2.85)	<0.001
75-84	2.21 (1.60, 3.06)	<0.001	2.97 (2.49, 3.54)	<0.001
85+	2.54 (1.78, 3.62)	<0.001	3.28 (2.71, 3.97)	<0.001
Female sex (Ref: Male)	1.08 (0.97, 1.19)	0.15	1.01 (0.95, 1.06)	0.85
Region (Ref: Northeast)
Midwest	0.91 (0.79, 1.05)	0.20	0.80 (0.74, 0.87)	<0.001
South	1.02 (0.89, 1.17)	0.79	1.00 (0.93, 1.07)	0.90
West	0.99 (0.82, 1.19)	0.91	0.83 (0.75, 0.92)	<0.001
Rural	0.86 (0.75, 0.99)	0.03	0.89 (0.82, 0.95)	0.001
Diagnoses of the lumbar spine
Radiculopathy	1.02 (0.91, 1.14)	0.73	1.00 (0.94, 1.07)	0.91
Stenosis	0.92 (0.82, 1.03)	0.14	0.97 (0.91, 1.03)	0.33
Herniated disc	1.08 (0.96, 1.20)	0.19	1.02 (0.96, 1.08)	0.59
Degenerative disc disease	0.93 (0.82, 1.05)	0.27	0.94 (0.88, 1.01)	0.08
Spondylosis	0.98 (0.83, 1.15)	0.79	0.93 (0.85, 1.01)	0.10
Myelopathy	1.25 (0.96, 1.63)	0.09	1.09 (0.94, 1.27)	0.25
Sciatica	1.08 (0.86, 1.34)	0.52	0.92 (0.91, 1.05)	0.22
Post-laminectomy syndrome	1.00 (0.79, 1.25)	0.97	1.03 (0.91, 1.16)	0.67
Other lumbar spine disorder	1.03 (0.85, 1.23)	0.78	1.13 (1.03, 1.25)	0.01
LESI details
Approach (Ref: Interlaminar)
Both	0.91 (0.56, 1.47)	0.70	0.78 (0.58, 1.03)	0.08
Transforaminal	0.82 (0.74, 0.90)	<0.001	1.01 (0.96, 1.06)	0.77
Injection year (Ref: 2014)
2015	1.00 (0.87, 1.15)	0.99	1.01 (0.94, 1.09)	0.71
2016	0.98 (0.84, 1.15)	0.83	0.97 (0.89, 1.06)	0.49
2017	1.00 (0.84, 1.19)	0.97	1.01 (0.92, 1.11)	0.87
2018	0.92 (0.75, 1.12)	0.42	1.01 (0.91, 1.12)	0.86
2019	1.11 (0.92, 1.34)	0.26	1.04 (0.94, 1.15)	0.41
2020	1.08 (0.87, 1.33)	0.48	1.00 (0.89, 1.12)	0.99
2021	1.04 (0.83, 1.29)	0.74	0.99 (0.88, 1.12)	0.88
Prior LESI in past year	0.95 (0.91, 0.99)	0.02	0.98 (0.96, 1.00)	0.10
Comorbidities
Depression	1.06 (0.93, 1.21)	0.36	1.06 (0.99, 1.14)	0.10
Anxiety	1.02 (0.90, 1.16)	0.72	1.11 (1.04, 1.19)	0.002
Other psychiatric conditions	1.39 (1.20, 1.61)	<0.001	1.22 (1.12, 1.32)	<0.001
COPD	0.98 (0.83, 1.16)	0.84	1.10 (1.01, 1.20)	0.02
HF	0.97 (0.81, 1.16)	0.74	0.98 (0.88, 1.08)	0.63
HTN	1.45 (1.28, 1.64)	<0.001	1.36 (1.27, 1.45)	<0.001
Previous MI	0.62 (0.45, 0.85)	0.003	0.97 (0.83, 1.12)	0.65
CAD	1.30 (1.14, 1.48)	<0.001	1.24 (1.16, 1.34)	<0.001
PAD	1.64 (1.43, 1.89)	<0.001	1.63 (1.51, 1.76)	<0.001
Anemia	0.98 (0.86, 1.12)	0.74	1.06 (0.99, 1.14)	0.10
Coagulation disorders	1.62 (1.21, 2.18)	0.001	1.36 (1.14, 1.64)	<0.001
Thrombocytopenia	1.27 (0.94, 1.73)	0.12	1.18 (0.99, 1.14)	0.07
Other cardiac conditions	1.78 (1.57, 2.02)	<0.001	1.53 (1.32, 1.64)	<0.001
Other pulmonary conditions	1.12 (0.98, 1.27)	0.09	1.06 (0.99, 1.14)	0.10
Diabetes	1.26 (1.13, 1.40)	<0.001	1.14 (1.08, 1.21)	<0.001
Renal dysfunction	1.33 (1.07, 1.64)	0.01	1.16 (1.02, 1.31)	0.02
Obesity	0.91 (0.81, 1.03)	0.15	0.99 (0.93, 1.06)	0.87
Alcohol use	0.84 (0.56, 1.25)	0.39	0.98 (0.80, 1.20)	0.87
Tobacco use	1.14 (0.98, 1.34)	0.09	1.26 (1.16, 1.37)	<0.001
Other drug use	1.18 (0.95, 1.46)	0.13	1.11 (0.99, 1.24)	0.09
Neurological disorders	2.67 (2.38, 3.00)	<0.001	2.29 (2.15, 2.44)	<0.001
Hyperlipidemia	1.33 (1.19, 1.48)	<0.001	1.30 (1.22, 1.38)	<0.001
GERD	0.97 (0.86, 1.09)	0.60	1.08 (1.02, 1.15)	0.01
Chronic pain	0.98 (0.88, 1.10)	0.76	1.10 (1.04, 1.16)	0.002
Immunodeficiency	0.95 (0.75, 1.21)	0.67	1.11 (0.98, 1.25)	0.10
Inflammatory disease	0.92 (0.76, 1.12)	0.42	1.00 (0.90, 1.10)	0.97
Steroid use	1.54 (0.98, 2.42)	0.06	1.02 (0.76, 1.36)	0.91
History of VTE	0.89 (0.50, 1.60)	0.70	0.79 (0.56, 1.12)	0.19
Medications
Anticoagulants	1.77 (1.47, 2.14)	<0.001	1.55 (1.39, 1.73)	<0.001
Antiplatelets	2.47 (2.10, 2.90)	<0.001	2.36 (2.16, 2.58)	<0.001
NSAIDs	0.94 (0.94, 1.07)	0.36	0.92 (0.86, 0.98)	0.01

For variables where a reference category is not specified, the comparator is yes vs no.

For the specific procedural complications of neurological injuries, increased age was associated with increased risk across both time points, with patients aged 65 and older having the greatest odds (P < 0.001 for all timepoints for each age group compared to ages 18-34) (Supplemental Table S2). Residence in rural areas was associated with decreased odds of neurological injury across all three timepoints [7 day OR: 0.86 (0.74, 0.99); P = 0.03]. The transforaminal method was associated with a lower risk of neurological injury compared to the interlaminar approach at 7 days [7 day OR: 0.81 (0.74, 0.90); P < 0.001]. The patient comorbidities with the highest associations for neurological injuries were previous neurological disorders [7 day OR: 2.77 (2.46, 3.12); P < 0.001], peripheral artery disease [7 day OR: 1.66 (1.44, 1.91); P < 0.001], hypertension [7 day OR: 1.44 (1.27, 1.64); P < 0.001], other cardiac conditions [7 day OR: 1.79 (1.58, 2.03); P < 0.001], and psychiatric conditions [7 day OR: 1.43 (1.23, 1.66); P < 0.001]. Antiplatelet [7 day OR: 2.62 (2.22, 3.08); P < 0.001] and anticoagulant [7 day OR: 1.81 (1.49, 2.19); P < 0.001] prescriptions had strong associations with neurological injuries (Supplemental Table S2).

For the specific procedural complication of epidural hematomas, the odds for epidural bleeds decreased in later years of the cohort. A history of peripheral artery disease was associated with increased odds of epidural hematomas at 30 days [30 day OR: 1.92 (1.07, 3.44); P = 0.03] (Supplemental Table S3). Renal dysfunction had the greatest association for epidural hematoma at 30 days [30 day OR: 2.67 (1.26, 5.62); P = 0.01]. Other risk factors include coronary artery disease, anemia, obesity, and other cardiac conditions (Supplemental Table S3).

Factors Associated with Medical Complications

Age was determined to be a predictor for medical complications, where patients aged 45 and above had the greatest odds after LESI procedures (P < 0.001 for both timepoints for each age group compared to ages 18-34) (Table 6). Residence in the West [7 day OR: 1.20 (1.10, 1.30); P < 0.001] and female sex [7 day OR: 1.14 (1.09, 1.19); P < 0.001] was associated with greater odds for medical complications at all three timepoints. The patient comorbidities with the highest associations for increased medical complication odds were congestive heart failure [7 day OR: 2.51 (2.35, 2.69); P < 0.001] and other cardiac disorders [7 day OR: 1.82 (1.72, 1.93); P < 0.001]. Other comorbidities with significant associations included anxiety, renal dysfunction, hypertension, chronic obstructive pulmonary disease, anemia, coagulation disorders, thrombocytopenia, obesity, and gastroesophageal reflux disease. Anticoagulant prescription had the strongest association with medical complications at 7 [4.61 (4.33, 4.92); P < 0.001] and 30 days [5.71 (5.47, 5.97); P < 0.001] after LESI procedures. Antiplatelets use was not related to medical complication odds. NSAID prescriptions were associated with decreased odds for medical complications at all timepoints [7 day OR: 0.87 (0.82, 0.93); P < 0.001] (Table 6).

Table 6.

Patient Demographics, Clinical Factors, and Injection Characteristics Associated With Medical Complications at 7 and 30 Days

	7 day		30 day
Characteristic	OR (95% CI)	P-value	OR (95% CI)	P-value
Age (Ref: 18-34)
35-44	1.14 (0.96, 1.34)	0.12	1.04 (0.95, 1.13)	0.41
45-54	1.20 (1.03, 1.40)	0.02	1.18 (1.09, 1.28)	<0.001
55-64	1.69 (1.46, 1.97)	<0.001	1.51 (1.40, 1.63)	<0.001
65-74	2.34 (2.00, 2.73)	<0.001	2.11 (1.94, 2.29)	<0.001
75-84	2.96 (2.52, 3.46)	<0.001	2.78 (2.56, 3.02)	<0.001
85+	3.64 (3.08, 4.31)	<0.001	3.49 (3.19, 3.82)	<0.001
Female sex (Ref: Male)	1.14 (1.09, 1.19)	<0.001	1.26 (1.23, 1.30)	<0.001
Region (Ref: Northeast)
Midwest	1.00 (0.94, 1.07)	0.99	0.94 (0.91, 0.98)	0.002
South	1.02 (0.96, 1.09)	0.50	0.98 (0.94, 1.02)	0.25
West	1.20 (1.10, 1.30)	<0.001	1.08 (1.03, 1.13)	0.002
Rural	1.02 (0.96, 1.08)	0.54	0.99 (0.96, 1.03)	0.67
Diagnoses of the lumbar spine
Radiculopathy	1.02 (0.97, 1.07)	0.43	1.00 (0.97, 1.03)	0.95
Stenosis	1.01 (0.96, 1.07)	0.62	1.02 (0.99, 1.05)	0.15
Herniated disc	0.96 (0.91, 1.02)	0.18	0.97 (0.94, 1.00)	0.03
Degenerative disc disease	1.03 (0.98, 1.09)	0.25	1.01 (0.98, 1.04)	0.69
Spondylosis	1.05 (0.97, 1.12)	0.23	1.03 (0.99, 1.07)	0.19
Myelopathy	0.92 (0.80, 1.05)	0.21	0.90 (0.84, 0.97)	0.008
Sciatica	1.05 (0.94, 1.16)	0.40	0.95 (0.90, 1.01)	0.12
Post-laminectomy syndrome	1.03 (0.93, 1.14)	0.52	1.03 (0.97, 1.10)	0.27
Other lumbar spine disorder	1.04 (0.96, 1.13)	0.30	1.03 (0.99, 1.08)	0.17
LESI details
Approach (Ref: Interlaminar)
Both	0.92 (0.73, 1.15)	0.45	0.94 (0.83, 1.07)	0.34
Transforaminal	0.97 (0.93, 1.02)	0.20	0.98 (0.95, 1.00)	0.05
Injection year (Ref: 2014)
2015	0.95 (0.90, 1.02)	0.15	0.96 (0.92, 0.99)	0.02
2016	0.85 (0.79, 0.91)	<0.001	0.87 (0.84, 0.91)	<0.001
2017	0.78 (0.72, 0.85)	<0.001	0.82 (0.79, 0.86)	<0.001
2018	0.77 (0.70, 0.85)	<0.001	0.83 (0.79, 0.87)	<0.001
2019	0.84 (0.77, 0.91)	<0.001	0.87 (0.83, 0.92)	<0.001
2020	0.79 (0.72, 0.87)	<0.001	0.84 (0.80, 0.89)	<0.001
2021	0.81 (0.73, 0.89)	<0.001	0.83 (0.78, 0.88)	<0.001
Prior LESI in past year	0.99 (0.97, 1.01)	0.19	1.01 (1.00, 1.02)	0.05
Comorbidities
Depression	1.04 (0.98, 1.11)	0.18	1.10 (1.06, 1.13)	<0.001
Anxiety	1.09 (1.03, 1.16)	0.003	1.13 (1.09, 1.16)	<0.001
Other psychiatric conditions	1.19 (1.11, 1.28)	<0.001	1.19 (1.14, 1.24)	<0.001
COPD	1.08 (1.01, 1.16)	0.03	1.17 (1.12, 1.22)	<0.001
HF	2.51 (2.35, 2.69)	<0.001	2.93 (2.81, 3.06)	<0.001
HTN	1.37 (1.29, 1.45)	<0.001	1.31 (1.27, 1.35)	<0.001
Previous MI	1.02 (0.91, 1.15)	0.72	1.12 (1.04, 1.21)	0.003
CAD	1.28 (1.20, 1.35)	<0.001	1.36 (1.25, 1.48)	<0.001
PAD	1.15 (1.07, 1.23)	<0.001	1.12 (1.07, 1.17)	<0.001
Anemia	1.39 (1.32, 1.47)	<0.001	1.30 (1.26, 1.34)	<0.001
Coagulation disorders	1.40 (1.23, 1.59)	<0.001	1.36 (1.25, 1.48)	<0.001
Thrombocytopenia	1.19 (1.04, 1.36)	0.01	1.19 (1.09, 1.30)	<0.001
Other cardiac conditions	1.82 (1.72, 1.93)	<0.001	1.78 (1.72, 1.83)	<0.001
Other pulmonary conditions	1.20 (1.13, 1.27)	<0.001	1.16 (1.13, 1.20)	<0.001
Diabetes	1.16 (1.11, 1.22)	<0.001	1.15 (1.11, 1.18)	<0.001
Renal dysfunction	1.69 (1.55, 1.84)	<0.001	1.69 (1.59, 1.79)	<0.001
Obesity	1.20 (1.14, 1.27)	<0.001	1.17 (1.14, 1.21)	<0.001
Alcohol use	1.16 (0.9, 1.37)	0.07	1.14 (1.03, 1.26)	0.009
Tobacco use	1.01 (0.94, 1.09)	0.70	0.99 (0.95, 1.03)	0.55
Other drug use	1.10 (0.99, 1.21)	0.07	1.07 (1.01, 1.13)	0.02
Neurological disorders	1.19 (1.12, 1.27)	<0.001	1.22 (1.17, 1.27)	<0.001
Hyperlipidemia	1.01 (0.96, 1.06)	0.69	1.03 (1.00, 1.06)	0.03
GERD	1.15 (1.09, 1.21)	<0.001	1.13 (1.09, 1.16)	<0.001
Chronic pain	1.11 (1.06, 1.17)	<0.001	1.12 (1.08, 1.15)	<0.001
Immunodeficiency	1.26 (1.15, 1.39)	<0.001	1.27 (1.20, 1.35)	<0.001
Inflammatory disease	1.06 (0.98, 1.16)	0.15	1.11 (1.06, 1.16)	<0.001
Steroid use	1.18 (0.96, 1.46)	0.12	1.13 (0.99, 1.29)	0.07
History of VTE	1.24 (1.02, 1.50)	0.03	1.57 (1.38, 1.79)	<0.001
Medications
Anticoagulants	4.61 (4.33, 4.92)	<0.001	5.71 (5.47, 5.97)	<0.001
Antiplatelets	0.94 (0.85, 1.04)	0.26	0.97 (0.91, 1.03)	0.27
NSAIDs	0.87 (0.82, 0.93)	<0.001	0.90 (0.87, 0.92)	<0.001

For variables where a reference category is not specified, the comparator is yes vs no.

Discussion

In this large retrospective study utilizing the Merative MarketScan database, we aimed to evaluate the incidence and identify potential risk factors for complications following LESIs. Our findings provide valuable insights into the frequency and nature of these complications. Notably, when comparing the two injection approaches, our study revealed that the transforaminal approach was associated with fewer short-term (0-7 day) procedural complications. The interlaminar approach, on the other hand, showed a higher incidence of short-term complications. Our study also reveals that multiple factors were associated with procedural, medical, and overall complications, including age, sex, anticoagulant prescription, heart failure, neurological disorders, and renal dysfunction.

Our findings align with and expand upon previous literature regarding complications following ESIs. For example, a systematic review conducted by Abdi et al reported a lower overall complication rate (0.7% to 4.7%) for ESIs compared to our study. However, their analysis did not separate complications by time point.²⁹ Similarly, Chou et al observed comparable complication trends in their systematic review analyzing the use of ESIs to manage spinal stenosis and radiculopathy.⁷ Their focus, though, was largely on the interlaminar approach, which they suggested might be linked to higher early complications such as vascular injury or dural puncture.⁷ In contrast, our study benefits from a larger cohort size, providing more robust statistical power. Additionally, we adjusted for a broader set of confounding variables, such as comorbidities, medications, and age, which allows for more accurate interpretation of the results, especially when compared to smaller studies. Supporting this, Lee et al reviewed the safety profile of epidural steroids and emphasized that while complications are generally uncommon and rarely serious, various systemic and local adverse effects can occur with procedural technique.³⁰ Our study also provides a more granular understanding of when these complications occur, offering better insights than previous work, which typically reported only aggregate complication rates.

Our analysis of the two injection techniques revealed key differences in complication patterns associated with the transforaminal and interlaminar approaches. The transforaminal approach was associated with a lower incidence of early procedural complications, which could be attributed to the higher precision of the technique in targeting the epidural space, thus minimizing trauma to surrounding structures like the dura and blood vessels. However, the transforaminal approach was linked to a higher rate of late complications, including nerve root injury and epidural hematomas, which may be a consequence of the deeper needle penetration, closer to the epidural vessels and the nerve root in Kambin’s triangle, required for this technique. Additionally, the interlaminar approach, which is more superficial, showed a higher rate of early complications like dural puncture but did not lead to significant long-term issues. This suggests that while the transforaminal approach may reduce risk of immediate complications, it carries a higher risk of complications that develop over time such as neurological injuries, albeit a very low risk. Existing literature supports our findings that the transforaminal approach can lead to more complications due to its proximity to nerve roots and the need for deeper needle placement. Kreiner et al noted that the transforaminal approach was linked to a higher rate of nerve root injury due to its proximity.³¹ Similarly, Epstein et al reviewed the complications of epidural steroid injections such as dural puncture and vascular injury, likely due to the anatomy involved and needle trajectory.³²

Our study also highlights how demographic factors are associated with complications after LESIs. Older age was consistently associated with increased risk across all complication types and timepoints. This likely reflects age-related physiological decline, higher prevalence of systemic comorbidities, and increased frailty amongst older individuals. Regional variation was also observed. Interestingly, residence in rural areas was associated with lower odds of procedural complications, which may suggest a strictly selected patient pool with administration of a conservatively offered procedural intensity. Given limited access to pain management and specialty care, patients in rural areas may be selectively referred for LESIs only when more conservative measures are exhausted, leading to a narrower and potentially healthier procedural cohort.³³ Slower adoption of more technically complex injection techniques in rural practices may also contribute to lower complication rates, as practitioners may rely more heavily on well-established approaches with known safety profiles.³⁴

Regarding comorbidities, cardiac comorbidities, especially heart failure, were among the most impactful predictors of medical complications. The presence of heart failure likely signifies reduced physiological health, placing patients at higher risk for fluid imbalance, arrhythmias, or decompensation following even minimally invasive procedures. Hypertension, coronary artery disease, and peripheral artery disease similarly contributed to elevated risk, reinforcing the relevance of underlying vascular dysfunction in LESI outcomes. Hematologic factors were also strongly correlated with procedural risk, as patients with coagulation disorders or those prescribed anticoagulants demonstrated markedly higher risk of complications. Epidural hematoma occurrence was demonstrated to have a higher odds of occurring with comorbid cardiac and hematological conditions. Vascular injury is an inherent risk in spinal injections, and when combined with impaired coagulation or antiplatelet use, the risk of hematoma formation increases.³⁵ Our findings reinforce the importance of strict adherence to peri-procedural guidelines. Warfarin, clopidogrel, low-molecular-weight heparin, and other agents must be appropriately managed before intervention, and close coordination with prescribing clinicians is essential. Renal dysfunction also increased the likelihood of complications, consistent with its association with impaired drug clearance, electrolyte instability, and increased infection risk. Neurological disorders, which encompassed a broad range of conditions including Alzheimer’s disease, Parkinson’s disease, multiple sclerosis, and others, were associated with increased odds of procedural complications. Patients with these conditions may be more susceptible due to pre-existing neurological deficits, which may make procedural targeting more complex and complications more difficult to recognize.

These findings have direct implications for clinical decision-making. The consistent rise in complication rates as the post-procedure time frame expanded from 7 days to 30 days suggests that the risk does not end with the LESI itself and careful post-procedural monitoring should extend well into the intermediate period, especially for patients with known risk factors. Existing safety guidelines emphasize procedural caution and awareness of rare but serious complications; our results affirm these priorities while highlighting a broader need to incorporate patient-specific medical risk into planning and follow-up. Stratifying patients based on demographics, comorbidities, and medication use may inform both procedural candidacy and approach selection, as well as guide counseling regarding delayed risks. Routine individualized risk assessment should therfore be an integral component of LESI decision-making.

Limitations

These findings should be considered in the context of several limitations. Although we report the incidence of complications following LESIs, such events may reflect underlying patient health or concurrent treatments and may not be causally linked to the procedure itself. Medical complications, such as heart failure exacerbation or urinary tract infections, may particularly reflect unrelated patient factors despite the temporal proximity to the LESI. Attributing complications to the LESI is further challenged by the lack of a comparator treatment group. Further, the study relies on administrative claims data, which are inherently subject to coding inaccuracies and potential misclassification of both procedures and complications. As an insurance claims database, MarketScan captures longitudinal encounters across outpatient, emergency, and inpatient settings, enabling detection of complications such as epidural hematoma even when patients present to a hospital distinct from the procedural site, although clinical detail and severity cannot be confirmed. This may lead to underreporting or misreporting of certain outcomes, especially rare or subclinical complications. Although the dataset includes a large and diverse population, it primarily captures individuals with employer-sponsored insurance, which may limit generalizability to uninsured or Medicaid populations. Moreover, the use of deidentified claims data precluded the ability to confirm clinical diagnoses, procedural technique specifics, or severity of complications via chart review. Residual confounding from unmeasured factors, including injection technique nuances, provider experience, or over-the-counter medication use, cannot be ruled out. Data regarding cessation of anti-platelet agents and anticoagulants are also not available through the database. Regarding LESI details, the database did not provide information regarding which medication was instilled into the epidural space on injection. Further, narratives and reasons for why which approach was performed was no specified. Whether the LESI was performed with imaging or no imaging was unable to be determined given the lack of determination in the database. Despite these limitations, our study’s strengths include the use of a large cohort of over 720,000 LESI procedures, detailed stratification of complication timing, and robust multivariable modeling.

Conclusions

This study identified an overall complication rate of 2.5% at 7 days and 8.8% at 30 days post-procedure. Procedural complications were less frequent than medical complications, with neurological injury being the most common. While the interlaminar approach was associated with a greater rate of early procedural complications, the transforaminal approach showed increased odds of procedural and overall complications with a longer post-LESI time course. Key patient-level risk factors for complications included advanced age, anticoagulant or antiplatelet use, heart failure, renal dysfunction, neurological disorders, and psychiatric comorbidities. The diverse span of risk factors identified from our findings underscore the importance of individualized risk stratification prior to LESI administration. Understanding the nuanced relationship between injection technique, patient demographics, and comorbidities can help clinicians minimize adverse outcomes from LESIs and guide informed decision-making for safe and effective management of lumbar spine pathology.

Supplemental Material

Supplemental material - Complications After Lumbar Spine Epidural Steroid Injections: Incidence and Risk Factors

Supplemental material for Complications After Lumbar Spine Epidural Steroid Injections: Incidence and Risk Factors in Luca M. Valdivia, MS, Mayuri Jain, MPH, Olgerta Mucollari, BS, Charu Jain, BA, Brocha Z. Stern, PhD, Saad B. Chaudhary, MD by Global Spine Journal

Footnotes

ORCID iD

Luca M. Valdivia

Ethical Considerations

IRB approval was not required for this manuscript as only de-identified data were included in the investigation.

Consent to Participate

Informed consent was not needed for this study, as only de-identified patient data was utilized.

Author Contributions

LMV – Conception & Design, Drafting of the Manuscript, Statistical Analysis

MJ – Acquisition of Data, Analysis and Interpretation of Data, Critical Revision of Manuscript, Statistical Analysis

OM – Conception and Design, Drafting of Manuscript, Statistical Analysis, Administrative & Technical Support

CJ – Statistical Analysis, Drafting of Manuscript

BZ – Conception and Design, Acquisition of Data, Analysis and Interpretation of Data, Critical Revision of Manuscript, Obtaining Funding, Supervision

SBC – Conception and Design, Critical Revision of Manuscript, Administrative & Technical Support, Supervision

Funding

The authors disclosed receipt of the following financial support for the research, authorship, and/or publication of this article: This work was supported in part through the computational resources and staff expertise provided by Scientific Computing and Data at the Icahn School of Medicine at Mount Sinai and supported by the Clinical and Translational Science Awards (CTSA) grant UL1TR004419 from the National Center for Advancing Translational Sciences.

Declaration of Conflicting Interests

The authors declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.

Data Availability Statement

Only de-identified data were included in the investigation from the Merative MarketScan Commercial Claims and Encounters and Medicare Supplemental database.*

Supplemental Material

Supplemental material is available online.

References

Friedly

Comstock

Turner

, et al. A randomized trial of epidural glucocorticoid injections for spinal stenosis. N Engl J Med. 2014;371:11-21. doi:10.1056/NEJMoa1313265. Published online July 3 2014.

Friedly

Chan

Deyo

. Increases in Lumbosacral injections in the medicare population: 1994 to 2001. Spine. 2007;32(16):1754-1760. doi:10.1097/BRS.0b013e3180b9f96e

Manchikanti

Pampati

Falco

Hirsch

. Assessment of the growth of epidural injections in the medicare population from 2000 to 2011. Pain Physician. 2013;16(4):E349, E364. Accessed July 3, 2025.https://pubmed.ncbi.nlm.nih.gov/23877459/

Manchikanti

Knezevic

Navani

, et al. Epidural interventions in the management of chronic spinal pain: American society of interventional pain physicians (ASIPP) comprehensive evidence-based guidelines. Pain Physician. 2021;24:S27, S208. Accessed July 3, 2025.https://pubmed.ncbi.nlm.nih.gov/33492918/

Oliveira

Maher

Ferreira

, et al. Epidural corticosteroid injections for Lumbosacral radicular pain. Cochrane Database Syst Rev. 2020;4(4):CD013577. doi:10.1002/14651858.CD013577

Harrast

. Epidural steroid injections for lumbar spinal stenosis. Current Reviews in Musculoskeletal Medicine. 2007;1(1):32-38.

Chou

Hashimoto

Friedly

, et al. Epidural corticosteroid injections for radiculopathy and spinal stenosis: a systematic review and meta-analysis. Ann Intern Med. 2015;163(5):373-381. doi:10.7326/M15-0934

Chang

. Complications associated with lumbar transforaminal epidural steroid injections. Curr Pain Headache Rep. 2020;24(11):1-6.

Manchikanti

Malla

Wargo

Cash

Pampati

Fellows

. A prospective evaluation of complications of 10,000 fluoroscopically directed epidural injections. Pain Physician. 2012;15(2):131-140. Accessed July 3, 2025.https://pubmed.ncbi.nlm.nih.gov/22430650/

10.

Karaman

Kavak

GÖ

Tüfek

Yldrm

. The complications of transforaminal lumbar epidural steroid injections. Spine. 2011;36(13):E819-E824.

11.

Botwin

Gruber

Bouchlas

Torres-Ramos

Freeman

Slaten

. Complications of fluoroscopically guided transforaminal lumbar epidural injections. Arch Phys Med Rehabil. 2000;81(8):1045-1050. doi:10.1053/apmr.2000.7166

12.

Manchikanti

Candido

Singh

, et al. Epidural steroid warning controversy still dogging FDA. Pain Physician. 2014;17(4):E451-E474. Accessed July 3, 2025.https://pubmed.ncbi.nlm.nih.gov/25054397/

13.

U.S. Food and Drug Administration . FDA drug safety communication: FDA requires label changes to warn of rare but serious neurologic problems after epidural corticosteroid injections for pain. 2014. https://www.fda.gov/drugs/drug-safety-and-availability/fda-drug-safety-communication-fda-requires-label-changes-warn-rare-serious-neurologic-problems-after. Accessed 10 July 2025.

14.

Young

. Transient blindness after lumbar epidural steroid injection: a case report and literature review. Spine. 2002;27(21):E476-E477. doi:10.1097/00007632-200211010-00022

15.

Tiso

Cutler

Catania

Whalen

. Adverse central nervous system sequelae after selective transforaminal block: the role of corticosteroids. Spine J : Official Journal of the North American Spine Society. 2004;4(4):468-474. doi:10.1016/j.spinee.2003.10.007

16.

Houten

Errico

. Paraplegia after Lumbosacral nerve root block: report of three cases. Spine J : Official Journal of the North American Spine Society. 2002;2(1):70-75. doi:10.1016/s1529-9430(01)00159-0

17.

Lyders

Morris

. A case of spinal cord infarction following lumbar transforaminal epidural steroid injection: MR imaging and angiographic findings. AJNR American journal of neuroradiology. 2009;30(9):1691-1693. doi:10.3174/ajnr.A1567

18.

Suresh

Berman

Connell

. Cerebellar and brainstem infarction as a complication of CT-guided transforaminal cervical nerve root block. Skelet Radiol. 2007;36(5):449-452. doi:10.1007/s00256-006-0215-0

19.

Rozin

Koehler

, et al. Death during transforaminal epidural steroid nerve root block (C7) due to perforation of the left vertebral artery. Am J Forensic Med Pathol. 2003;24(4):351-355. doi:10.1097/01.paf.0000097790.45455.45

20.

Kabbara

Rosenberg

Untal

. Methicillin-resistant Staphylococcus aureus epidural abscess after transforaminal epidural steroid injection. Pain Physician. 2004;7(2):269-272. Accessed July 3, 2025.https://pubmed.ncbi.nlm.nih.gov/16868603/

21.

Kim

Lee

Kim

Cho

. Spinal epidural hematoma occurring at a distance from the transforaminal epidural injection site: a case report. Medicine. 2019;98(30):e16654. doi:10.1097/MD.0000000000016654

22.

Gungor

Aiyer

. Epidural hematoma development contralateral to dura after lumbar transforaminal epidural steroid injection. Pain Manag. 2017;7(5):367-375. doi:10.2217/pmt-2017-0012

23.

Horlocker

Wedel

Benzon

, et al. Regional anesthesia in the anticoagulated patient: defining the risks (the second ASRA consensus conference on neuraxial anesthesia and anticoagulation). Reg Anesth Pain Med. 2003;28(3):172-197. doi:10.1053/rapm.2003.50046

24.

Cohen

Greuber

Vought

Lissin

. Safety of epidural steroid injections for Lumbosacral radicular pain: unmet medical need. Clin J Pain. 2021;37(9):707-717. doi:10.1097/AJP.0000000000000963

25.

Merative MarketScan research databases. Accessed July 3, 2025.https://www.merative.com/documents/merative-marketscan-research-databases

26.

Wadhwa

Rodhe

Koltsov

JCB

, et al. Incidence and risk factors for complications following cervical epidural steroid injections. Spine J. 2025;25(5):886-902.

27.

Elixhauser

Steiner

Harris

Coffey

. Comorbidity measures for use with administrative data. Med Care. 1998;36(1):8-27. doi:10.1097/00005650-199801000-00004

28.

Ratliff

Balise

Veeravagu

, et al. Predicting occurrence of spine surgery complications using “Big Data” modeling of an administrative claims database. J Bone Jt Surg Am Vol. 2016;98(10):824-834. doi:10.2106/JBJS.15.00301

29.

Abdi

Datta

Trescot

, et al. Epidural steroids in the management of chronic spinal pain: a systematic review. Pain Physician. 2007;10(1):185, 212. Accessed July 29, 2025.https://pubmed.ncbi.nlm.nih.gov/17256030/

30.

Lee

Moon

. Safety of epidural steroids: a review. Anesthesia and pain medicine. 2021;16(1):16-27. doi:10.17085/apm.21002

31.

Kreiner

Hwang

Easa

, et al. An evidence-based clinical guideline for the diagnosis and treatment of lumbar disc herniation with radiculopathy. Spine J : Official Journal of the North American Spine Society. 2014;14(1):180-191. doi:10.1016/j.spinee.2013.08.003

32.

Epstein

. Major risks and complications of cervical epidural steroid injections: an updated review. Surg Neurol Int. 2018;9:86.

33.

Baker

Liu

Bully

, et al. Overcoming barriers: a comprehensive review of chronic pain management and accessibility challenges in rural America. Health Care. 2024;12(17):1765. doi:10.3390/healthcare12171765

34.

Sticca

Burchill

Johnson

. Advanced technology and the rural surgeon. Surg Clin. 2020;100(5):909-920.

35.

Goodman

Posecion

LWF

Mallempati

Bayazitoglu

. Complications and pitfalls of lumbar interlaminar and transforaminal epidural injections. Curr Rev Musculoskelet Med. 2008;1(3-4):212-222.

Supplementary Material

Please find the following supplemental material available below.

For Open Access articles published under a Creative Commons License, all supplemental material carries the same license as the article it is associated with.

For non-Open Access articles published, all supplemental material carries a non-exclusive license, and permission requests for re-use of supplemental material or any part of supplemental material shall be sent directly to the copyright owner as specified in the copyright notice associated with the article.

0.15 MB

0.00 MB