Regional Anesthesia: Brief Report

The Risk and Outcomes of Epidural Hematomas After Perioperative and Obstetric Epidural Catheterization

A Report from the Multicenter Perioperative Outcomes Group Research Consortium

Bateman, Brian T. MD^*; Mhyre, Jill M. MD^†; Ehrenfeld, Jesse MD, MPH^‡; Kheterpal, Sachin MD, MBA^†; Abbey, Kenneth R. MD, JD^§; Argalious, Maged MD, MBA^‖; Berman, Mitchell F. MD, MPH^¶; Jacques, Paul St. MD^#; Levy, Warren MD^**; Loeb, Robert G. MD^††; Paganelli, William MD, PhD^‡‡; Smith, Kelly W. MD^§§; Wethington, Kevin L. MD^§§; Wax, David MD^‖‖; Pace, Nathan L. MD, MStat^§§; Tremper, Kevin MD, PhD^†; Sandberg, Warren S. MD, PhD^#

Author Information

From the ^*Department of Anesthesia, Critical Care, and Pain Medicine, Massachusetts General Hospital, and Division of Pharmacoepidemiology and Pharmacoeconomics, Brigham and Women’s Hospital, Harvard Medical School, Boston, Massachusetts; ^†Department of Anesthesiology, University of Michigan Health System, Ann Arbor, Michigan; ^‡Department of Anesthesiology and Biomedical Informatics, Vanderbilt University, Nashville, Tennessee; ^§Department of Anesthesiology and Perioperative Medicine, Oregon Health and Science University, Portland, Oregon; ^‖General Anesthesiology, Cleveland Clinic, Cleveland, Ohio; ^¶Department of Anesthesiology, Columbia University, New York, New York; ^#Departments of Anesthesiology, Biomedical Informatics, and Surgery, Vanderbilt University School of Medicine, Nashville, Tennessee; ^**Department of Anesthesia, Perelman School of Medicine at the University of Pennsylvania, Philadelphia, Pennsylvania; ^††Department of Anesthesiology, University of Arizona, Tucson, Arizona; ^‡‡Anesthesiology, Fletcher Allen Health Care—University of Vermont, Burlington, Vermont; ^§§Department of Anesthesiology, University of Utah, Salt Lake City, Utah; and ^‖‖Department of Anesthesiology, Mount Sinai School of Medicine, New York, New York.

Accepted for publication February 1, 2012.

Published ahead of print April 13, 2012

Supported by departmental funds.

Supplemental digital content is available for this article. Direct URL citations appear in the printed text and are provided in the HTML and PDF versions of this article on the journal’s Web site (www.anesthesia-analgesia.org).

The authors declare no conflicts of interest.

Reprints will not be available from the authors.

Address correspondence to Brian T. Bateman, MD, Department of Anesthesia, Critical Care, and Pain Medicine, Massachusetts General Hospital, Harvard Medical School, 55 Fruit St., Boston, MA 02114. Address e-mail to [email protected].

Anesthesia & Analgesia 116(6):p 1380-1385, June 2013. | DOI: 10.1213/ANE.0b013e318251daed

Free
SDC

BACKGROUND:

In this study, we sought to determine the frequency and outcomes of epidural hematomas after epidural catheterization.

METHODS:

Eleven centers participating in the Multicenter Perioperative Outcomes Group used electronic anesthesia information systems and quality assurance databases to identify patients who had epidural catheters inserted for either obstetrical or surgical indications. From this cohort, patients undergoing laminectomy for the evacuation of hematoma within 6 weeks of epidural placement were identified.

RESULTS:

Seven of 62,450 patients undergoing perioperative epidural catheterizations developed hematoma requiring surgical evacuation. The event rate was 11.2 × 10⁻⁵ (95% confidence interval [CI], 4.5 × 10⁻⁵ to 23.1 × 10⁻⁵). Four of the 7 had anticoagulation/antiplatelet therapy that deviated from American Society of Regional Anesthesia guidelines. None of 79,837 obstetric patients with epidural catheterizations developed hematoma (upper limit of the 95% CI, 4.6 × 10⁻⁵). The hematoma rate in obstetric epidural catheterizations was significantly lower than in perioperative epidural catheterizations (P = 0.003).

CONCLUSIONS:

In this series, the 95% CI for the frequency of epidural hematoma requiring laminectomy after epidural catheter placement for perioperative anesthesia/analgesia was 1 event per 22,189 placements to 1 event per 4330 placements. Risk was significantly lower in obstetric epidurals.

Epidural hematoma is a recognized complication of epidural catheterization. Recent studies examining this complication have been generally limited by being single-center, survey-based, lacking clear denominators, derived from European rather than North American experience, or having incomplete clinical detail about the affected patients.^1–5 The Multicenter Perioperative Outcomes Group (MPOG) is a consortium of academic anesthesia departments with electronic Anesthesia Information Management Systems (AIMS). MPOG pools perioperative data for research purposes. The consortium studied the event rate for epidural hematomas requiring laminectomy after epidural catheterization for perioperative or obstetric anesthesia/analgesia.

METHODS

This study was reviewed and approved by the IRBs of each of the contributing institutions. The study was a retrospective, observational study conducted at academic medical centers that are part of the MPOG consortium. Eleven institutions contributed data regarding epidural catheterization for perioperative anesthesia/analgesia and 6 institutions contributed data regarding obstetric epidural catheterization. Details regarding the characteristics of the institutions participating in the study are available in Online Appendix 1 (see Supplemental Digital Content 1, https://links.lww.com/AA/A382).

Investigators queried their AIMS and quality assurance databases for all epidural catheter insertions in operative patients or obstetric labor patients. The study dates varied by institution, depending on when deployment of their AIMS or quality assurance databases was complete. The cohort of all patients who underwent epidural catheterization was then automatically screened for patients undergoing operations within 6 weeks of the catheter placement using AIMS and/or billing records. This subset of patients and secondary operations was reviewed to identify patients undergoing a laminectomy for the evacuation of epidural hematoma.

Information from each patient requiring a decompressive laminectomy, including demographic characteristics, comorbidities, details of epidural catheterization, pre-, intra-, and postoperative coagulation variables, and antiplatelet and anticoagulant medication usage, were manually collected through use of electronic and paper medical chart review. Data were also collected on the presenting signs and symptoms of the hematoma, imaging findings, time to imaging and laminectomy, and neurological outcome. The data collection form is available in Online Appendix 2 (see Supplemental Digital Content 2, https://links.lww.com/AA/A383).

The point estimate and 95% confidence interval (CI) on the summary event rate across institutions was calculated using the Pearson-Clopper method. Comparisons of the risk between perioperative and obstetric epidural use were performed using the Fisher exact test; a 95% CI on the difference of proportions was also estimated. Calculations were performed using the binom package version 1.0-5 running in R version 2.14.0 (R: A Language and Environment for Statistical Computing; R Foundation for Statistical Computing, Vienna, Austria) and StatXact 9 (Cytel Inc., Cambridge, MA). As a sensitivity analysis, random effects meta-analysis and Bayesian models were obtained (Online Appendix 3, see Supplemental Digital Content 3, https://links.lww.com/AA/A384).

RESULTS

Six institutions reported a total of 79,837 obstetric epidural placements. There were no reported cases of hematoma requiring decompressive laminectomy for obstetric patients. Eleven institutions reported 62,450 epidural catheters inserted for perioperative anesthesia and analgesia. Among these patients, there were 7 decompressive laminectomies for hematoma for an unweighted risk of 1 laminectomy for hematoma per 8921 epidural placements. One institution had 2 events, 5 institutions had 1 event each, and 5 institutions had no events (Table 1).

T1-29 — Table 1:
The Total Number of Perioperative Epidural Placements and Laminectomies for Hematoma, by Institution

Patient characteristics, details regarding epidural catheterization, and information on perioperative anticoagulant/antiplatelet exposure for each case of hematoma requiring laminectomy are reported in Table 2. Five of the 7 epidurals were thoracic, the other 2 were lumbar. One case occurred in the setting of a combined spinal/epidural, another followed accidental dural puncture and placement of a spinal catheter. In all cases, the neuraxial catheter was in situ at the time that signs and symptoms of epidural hematoma were recognized. In 2 instances, the epidural catheters were inserted in the setting of elevated international normalized ratio (1.6 in both cases). In 5 cases, the patients received standard heparin in the perioperative period, and in 4 cases, the patients received aspirin.

T2-29 — Table 2-a:
Demographic and Clinical Characteristics of Patients Who Developed Epidural Hematomas Requiring Decompressive Laminectomy After Epidural Catheterization

T3-29 — Table 2-b:
Demographic and Clinical Characteristics of Patients Who Developed Epidural Hematomas Requiring Decompressive Laminectomy After Epidural Catheterization

Presenting symptoms, timing of the symptoms after epidural catheterization, timing of imaging and laminectomy after first symptoms, and neurological outcome are noted in Table 3. Motor deficits in the lower extremity were present in all 7 cases. The time to first symptoms after placement ranged from 11 to 71 hours. The timing of laminectomy after symptoms varied widely from 6.5 to 54 hours. There was no apparent relationship between the time to laminectomy and neurological recovery (in the 2 cases of complete recovery, laminectomy occurred at 7 hours and 54 hours after symptoms).

T4-29 — Table 3:
Clinical Course of Patients Who Developed Epidural Hematomas Requiring Decompressive Laminectomy After Epidural Catheterization

The risk of decompressive laminectomy for epidural hematoma among perioperative epidural catheter placement was 11.2 × 10⁻⁵ (95% CI, 4.5 × 10⁻⁵ to 23.1 × 10⁻⁵); the lower and upper bounds may also be stated as 1 event in 22,189 and 1 event in 4330 epidural placements. The risk of decompressive laminectomies for epidural hematoma among obstetric epidural catheter placements was 0 (95% CI, 0 to 4.6 × 10⁻⁵); the upper bound may be stated as 1 event in 21,643 epidural placements. The risk difference between perioperative and obstetric epidural catheter placement was also 11.2 × 10⁻⁵ (95% CI, 5.4 × 10⁻⁵ to 23.1 × 10⁻⁵) and was significant (Fisher exact test, P = 0.003).

DISCUSSION

We report herein the risk and outcomes of epidural hematomas requiring decompressive laminectomy from a large series of perioperative and obstetric epidural catheterizations at 11 academic medical centers. The unweighted event rate of epidural hematoma requiring laminectomy that we report is 1 per 8921 epidural catheterizations for perioperative anesthesia/analgesia. Five of the 7 patients with hematoma had neurological deficits on discharge, so the rate of neurological injury is approximately 1 in 12,000 epidural catheterizations. These estimates are comparable with other studies.^1,2,4 In the obstetric series, we report 79,837 epidural catheterizations without a single case of hematoma requiring laminectomy.

Although the number of cases of hematoma in our series is small, several interesting observations emerge. Patient 1 may have been at increased risk because of the deleterious effect of dialysis on platelet function. Four of 7 patients had perioperative anticoagulant management that clearly deviated from current American Society of Regional Anesthesia guidelines.⁶ Patients 2 and 7 had placements with international normalized ratios of 1.6. Patients 3, 6, and 7 all received concurrent aspirin and heparin. The final 2 cases (patients 4 and 5) both occurred in the setting of a dural puncture: one after combined spinal/epidural (after difficult placement in a patient with ankylosing spondylitis) and a second in the setting of an inadvertent dural puncture and spinal catheter. The time to development of symptoms after catheter insertion varied widely from 11 to 71 hours, with 5 of 7 cases presenting beyond the first 24 hours. This underscores the need for vigilance in monitoring for signs and symptoms of hematoma as long as the catheter is in place. In contrast to previous analyses,⁷ we observed no relationship between time to laminectomy and neurological outcome in our small sample. Although it is clearly prudent to proceed to decompression as rapidly as possible once the diagnosis is made, predictors of neurological recovery are likely multifactorial.

Despite the inclusion of 79,837 obstetric epidurals in our study, we did not detect a single case of hematoma requiring laminectomy, and the risk was significantly lower than that observed in perioperative epidural catheter placement. It is possible that these patients are at decreased risk compared with the general population because of the relatively hypercoagulable state associated with pregnancy. Obstetric patients are also less likely to have canal or foraminal stenoses or receive a combination of antiplatelet and anticoagulant drugs (although in certain situations, including history of venous thromboembolism, antiphospholipid antibody syndrome, and mechanical heart valves, anticoagulation during pregnancy may be required).⁸

Our study has certain limitations. Because our methods only detect patients who underwent decompressive laminectomies, we do not identify those hematomas that were managed non-operatively. Our data provide only a lower bound of the true risk, because patients conceivably could be discharged, develop symptoms, and undergo treatment at another hospital. Our data on neurological outcome are limited to that which is documented at the time of hospital discharge and therefore may overestimate the degree of ultimate neurological impairment. We did not collect information of attempted epidural placements that were aborted because of bleeding or difficult placement; therefore, we might have underestimated the risk of hematoma associated with attempted placement. Likewise, information was not uniformly available for all patients on the type of epidural catheters used and whether spinal stenosis was present.

In conclusion, epidural hematoma is a rare but serious complication after epidural catheterization. In our series, the frequency of epidural hematoma requiring laminectomy after epidural catheterization instituted for perioperative anesthesia/analgesia was between 1 event per 22,189 and 1 event per 4330 epidural catheter placements. In addition, the risk was significantly lower in obstetric patients.

DISCLOSURES

Name: Brian T. Bateman, MD.

Contribution: This author helped design the study, conduct the study, analyze the data, and prepare the manuscript.

Name: Jill M. Mhyre, MD.

Contribution: This author helped design the study, conduct the study, analyze the data, and prepare the manuscript.

Name: Jesse Ehrenfeld, MD, MPH.

Contribution: This author helped design the study, conduct the study, analyze the data, and prepare the manuscript.

Name: Sachin Kheterpal, MD, MBA.

Contribution: This author helped design the study, conduct the study, analyze the data, and prepare the manuscript.

Name: Kenneth R. Abbey, MD, JD.

Contribution: This author helped design the study, conduct the study, and prepare the manuscript.

Name: Maged Argalious, MD, MBA.