Introduction

Background

Pulmonary embolism (PE) is a potentially life threatening condition and a significant cause of morbidity and mortality in patients with spinal cord injury (SCI).1 Diagnosing PE in SCI patients can be challenging as usual clinical signs and symptoms of pulmonary emboli may be obscured.2

Diagnostic imaging of PE first involves a chest X-ray to rule out any conditions that mimic PE, including lobar pneumonia or pneumothorax. More definitive imaging investigations include computed tomography pulmonary angiography (CTPA)3 and less commonly, ventilation-perfusion scintigraphy.4 Multislice CTPA has become the principal modality for diagnosis as it is both highly sensitive and specific.3

In the acute phase of SCI (up to 90 days post injury) the incidence of PE has been reported as 4.5% with a mortality rate of 3.5%.5 The introduction of prophylactic anticoagulation therapy in the first 3 months was associated with a reduction in mortality rates.6 Anticoagulation is usually ceased after 3 months post injury, based on autopsy evidence of rare incidences of PE in this period.7, 8 However, the incidence of PE after the initial 3 months post injury is poorly reported in literature.9 In this study, over a 6-year period we examine the incidence rates of PE in SCI patients after anticoagulation was stopped.

Methods

Study design

This is a single-centre retrospective study of SCI. The study reviewed all patients with suspected PE who had a CTPA or ventilation-perfusion lung scan.

Setting

The study setting is the National Spinal Injuries Centre (NSIC) at Stoke Mandeville Hospital, Aylesbury, UK.

Participants

We included all new-onset SCI patients admitted between 2008 and 2013. We excluded children (under 18 years).

Outcomes

Primary outcome

Incidence of PE after 90 days post SCI.

Secondary outcomes.

  1. 1

    Duration of injury associated with PE.

  2. 2

    Extent of paralysis associated with PE.

  3. 3

    Clinical presentation of PE in SCI.

Data sources and management

All CTPA and perfusion scans were identified via Centricity Radiology Information Systems. All images were reviewed by an experienced consultant radiologist. The medical notes of patients with a suspected PE were reviewed and clinical outcomes and radiological findings were recorded. Microsoft Excel software (Microsoft Excel 2013, Redmond, WA, USA) was used to input admission information and collect data. Statistical analyses were performed using IBM SPSS (Version 22.0, Armonk, NY, USA) software.

Statistical methods

Incidence rate

Incidence rates post 90 days of SCI were calculated by dividing the number of new PEs by the number of new admissions in a year.

Follow-up

The follow-up of each individual patient lasted for the inpatient period only. Patients were not followed up after discharge.

Results

This study includes 640 patients with new-onet SCI. The mean injury to NSIC admission duration was 54 days; 95% confidence interval (CI) (50–57). The mean length of hospital stay was 150 days; 95% CI (144–155). The population baseline characteristics are found in Table 1.

Table 1 Study characteristics
Full size table

During the years from 2008 to 2013, 91 patients had a clinical suspicion for PE and subsequently had a CTPA or perfusion lung scan conducted. Eight patients had a confirmed PE 90 days post SCI. The mean duration of injury at the time of PE was 226 days; 95% CI (138.3–314.9).

The incidence of PE after 90 days post SCI was 1.25%; 95% CI (0.39–2.11) (Table 2).

Table 2 Incidence of pulmonary embolism 90 days post spinal injury
Full size table

Four patients of the eight patients diagnosed with PE had tetrapelgia. Five patients had a complete injury (Table 3).

Table 3 Demographic data of patients with PE
Full size table

The most common clinical presentations of PE were dyspnoea, chest pain, hypoxia, tachycardia and haemoptysis. Other presentations included atrial fibrillation, respiratory failure and cardiac arrest.

Discussion

Our analysis of pulmonary embolisms 90 days after SCI identified an incidence rate of 1.25%. The duration of SCI at the time of PE was 7 months.

The incidence of PE after 90 days post SCI is lower than the 5–12% incidence rate during the acute phase (first 3 months) as reported in the literature.1, 5 However, it is significantly higher than the PE risk in the general population (0.06% per year).10

Patients with a complete injury AIS A had a slightly higher rate of PE. There was no difference in PE rates between individuals with tetraplegia or paraplegia.

Conclusion

The risk of pulmonary embolism 90 days post SCI remains significant.

Limitations

This is a retrospective analysis of available SCI data at the National Spinal Injuries Centre. To be able to calculate an accurate incidence rate, a prospective multicentre cohort study would be more appropriate. It was beyond the scope of this study to determine correlation of pulmonary embolism with level of injury.

Data archiving

There were no data to deposit.