Electrocardiographic Findings in Chronic Spinal Cord Injury

Wilsa MS Charles, MA; B. Jenny Kiratli PhD; Jon N. Myers, PhD; Inder Perkash, MD; VF Froelicher, MD

Objective: The purpose of this study was to determine the prevalence and prognostic value of ECG abnormalities in patients with spinal cord injury (SCI) compared with the able-bodied (AB) population. Rehabilitation of patients with SCI includes prevention and management of diseases that can be accelerated by their injury. Cardiovascular morbidity and mortality are thought to be high in these patients due to sedentary lifestyle, poor aerobic fitness and higher occurrence of other risk factors including hypertension, hyperlipidemia, obesity, and diabetes. However, there is little evidence available regarding differences in prevalence, manifestations, and identification of heart disease in this population. Further, while the ECG is an inexpensive, reliable tool for assessment of heart disease, there are no established guidelines nor evidence available to support the inclusion of an ECG as a part of the mandated annual evaluation for patients with SCI.

Methods: An analysis of previously collected screening ECG studies was performed. Since 1987, all ECGs obtained at the VA Palo Alto have been digitally recorded and stored. The earliest ECG available in the database for each patient was used, and computerized measurements and interpretations were downloaded. ECG data were excluded during inpatient admissions. ECGs were available for 654 patients with SCI (mean age 50 14 years), and relevant demographic and injury-related information was recorded. Patients with SCI were categorized by level of injury at or above T5 (sympathetic innervation interrupted) versus at or below T6 (sympathetic innervation intact). ECG data from 26,734 able-bodied male veterans (55.8 15 years) were included for comparison. Subsequent deaths were determined and recorded for both populations.

Results: Little difference in prevalence of ECG abnormalities was noted in patients with SCI attributable to lesion level, except there were fewer inferior Q waves, increased left atrial abnormality, and greater ST elevation in those with interrupted sympathetic innervation. As expected, increases in prevalence were noted with age (patients with SCI older than 65 compared with those who were younger) for the majority of ECG abnormalities. However, prevalence patterns of patients with SCI were similar to those observed in able-bodied patients. Survival analysis was performed and significant hazard ratios noted only for RBBB, atrial fibrillation, and IVCD in patients with SCI while all ECG abnormalities generated significant hazard ratios in the able-bodied sample.

Conclusions: Prevalence of ECG abnormalities in patients with SCI were similar to those observed in the reference non-injured population, suggesting that the ECG has similar value as a screening tool in the two populations. Further, although hazard ratios were generally higher in the able-bodied population, this may be somewhat attributable to different endpoints used (cardiac death only in patients with SCI compared with all-cause mortality the able-bodied sample). In summary, the prevalence of ECG findings in patients with SCI provides support for the ECG as part of routine/annual clinical evaluation in patients with SCI, coupled with assessment of concurrent cardiac disease. Evaluation of follow-up ECGs and prospective evaluation of cardiac mortality, currently underway, is expected to strengthen the evidence for prognostic interpretation of ECG findings in this population.

Acknowledgment: This study was funded by the VA Rehabilitation Research and Development Service, project # B2140-R.