Introduction
Interspinous bursitis has been reported to occur in both the cervical and lumbar spine, with the latter eponymously known as Baastrup's disease [1]. Despite anatomical studies demonstrating the prevalence of these bursae to be up to 50% [2], the clinical course is poorly understood, with radiographical findings not frequently correlating with patient symptoms [3]. This poses a challenge to the treating physician and requires a high degree of suspicion to prevent an erroneous diagnosis.
We present the case of a 13‐year‐old male, who attended the emergency department with a 1‐month history of atraumatic pain at the junction of his cervical and thoracic spine. We document the clinical course and investigations required to make the diagnosis of interspinous bursitis. Of note, this exercise‐associated presentation is, to the best of our knowledge, the first documented pediatric case in the literature.
Case History/Examination
Our patient presented to the emergency department with a 1‐month history of pain. This pain was most pronounced at his vertebrae prominens. He experienced no prodromal symptoms and had no prior spinal pathology.
He was previously fit and well with no known medical history. He denied any associated appendicular paresthesia or subjective weakness. He reported normal bowel and bladder function before and after the onset of symptoms.
This cervical spine pain coincided with the onset of swimming lessons when our patient was learning to perform the butterfly stroke. He denied prior episodes of pain in his axial skeleton or a history of trauma. His symptoms were exacerbated by carrying his school bag, and he achieved minimal relief from rest.
Our patient had a normal neurological examination of his upper and lower limbs throughout his clinical course. He exhibited full power in all upper and lower limb myotomes. His pinprick and soft‐touch sensation was intact throughout all dermatomes.
His biceps, triceps, quadriceps, and Achilles tendon reflexes were normal with a grading of 2+ bilaterally. Babinski and Hoffmann's reflexes were normal, and at no time point was clonus evident. His Romberg's test was normal, and he demonstrated a normal tandem gait.
He was focally tender at the junction of the thoracic and cervical spine with no overlying soft‐tissue changes. He had reduced range of motion on flexion and extension in his cervical spine, with both limited to 20 degrees secondary to pain. His Spurling test was negative.
Differential Diagnosis
Given the nonspecific findings on history and examination, our differential diagnosis remained broad. Sinister causes including malignancy and infection were considered. More common etiologies such as musculoskeletal pain and rheumatological entities were deemed to be more likely during the workup of these symptoms.
Due to persisting symptoms despite a period of rest, a non‐contrast magnetic resonance imaging (MRI) scan was performed of the thoracic and cervical spine. This demonstrated inflammatory changes at the tip of the C7 and T1 spinous processes, with findings in keeping with cervical interspinous bursitis. Edema was noted at the tip of the spinous processes with an associated small fluid volume.
Edema was also noted to be present in the adjacent soft tissues. These findings were most evident on short tau inversion recovery (STIR) sequences shown in Figure 1. Subsequent plain films noted loss of cervical lordosis but were otherwise unremarkable.
These are presented in Figure 2.
Given the differential diagnosis of chronic recurrent multifocal osteomyelitis, the rheumatology service was consulted. A serological workup yielded no evidence of active rheumatological disease. Results included a C‐reactive protein < 4 mg/L, erythrocyte sedimentation rate 2 mm/h, negative rheumatoid factor, anti‐streptolysin titer 52 IU/mL, and a viral screen for Epstein–Barr, cytomegalovirus, and tuberculosis all reported as negative.
Thereafter, a non‐contrast whole‐body MRI was completed 7 days following the index scan. This was in keeping with the aforementioned persistent interspinous bursitis at C7 and T1, with no other focal abnormality demonstrated.
Following the consultation to the spinal service, the consensus for management was nonoperative. This took the form of immobilization in a Miami‐J collar (Össur Headquarters, Reykjavik, Iceland) for comfort in combination with anti‐inflammatory oral analgesics. The purposes of this collar were solely for analgesic purposes and were not used as a component of the diagnostic strategy.
Pain was further controlled with oral paracetamol and ibuprofen. At no point was opiate analgesia required. Our patient reported improvements in subjective pain scoring with this regular analgesia, and the collar was subsequently weaned by the time of discharge on Day 7 post‐admission.
Our patient remained as an inpatient for both analgesic and diagnostic workup purposes. Given the broad differential diagnosis on presentation, it was deemed prudent to complete this diagnostic process in full as an inpatient. It was not pain management nor was an alternative clinical concern that was a contributing factor to this 7‐day admission, but the combination of whole‐body MRI scan and complete serological results was.
Conclusion and Results
Our patient was followed up in the outpatient setting, and he underwent a repeat MRI scan at 3 months post‐onset of symptoms. Interval improvement in previous bone marrow edema with a small volume of residual STIR signal at the junction of C7 and T1 was observed. These updated findings are displayed in Figure 3.
At the clinical review 4 months post‐discharge, our patient had resolution of his symptoms. He reported no pain during daily activities and had ceased his oral analgesics. On examination, he had mild residual tenderness on palpation of the spinous processes at the lower cervical spine.
The neurological examination of his upper and lower limbs remained unremarkable. Overall, this clinical course lasted for 4 months. It began 1 month prior to acute presentation to hospital, followed by an inpatient admission for investigations and treatment lasting 7 days.
Final follow‐up was completed at 3 months after discharge, with a full recovery achieved in the intervening months.
Discussion
First described by Baastrup in 1933 [4], interspinous bursitis has been described as a rare pathology of the lumbar spine. He noted that spinous processes could be driven toward one another when the lumbar spine is hyper‐lordotic. This mechanical impingement of the spinous processes has given rise to the more colloquial term “kissing spine disease.” Baastrup's disease is more commonly found in older adults, with management options ranging from physiotherapy up to surgical decompression [5].
Literature for pediatric patients experiencing Baastrup's disease is sparse, with only five case reports noted to date [6]. All of these previously described cases involved the lower thoracic or lumbar spine, with four occurring in gymnasts.
Although extremely rare, cervical interspinous bursitis may be associated with polymyalgia rheumatica (PMR) [7, 8]. Interestingly, Salvarani et al. [9] reported their findings in 12 adult patients with PMR, all of whom were found to have cervical interspinous bursitis on their MRI scans. Furthermore, a postmortem series from Bywaters in 1982 previously demonstrated cervical interspinous bursae in 14 out of 27 “normal” adult cervical spines [2].
We propose that swimming using the butterfly stroke may have been the instigating factor for this episode of cervical spine bursitis. The butterfly stroke requires recurrent spine hyperextension, followed by hyperflexion to create a whip‐like kick, providing forward momentum [10]. Our patient was not a regular swimmer prior to this episode, nor did he partake in sports on a regular basis.
This sudden increase in physical activity combined with the rhythmic motion required for the butterfly stroke may have predisposed him to developing interspinous bursitis. We acknowledge that we are unable to objectively prove this causation. This association is hypothetical based on the temporal correlation with the onset of symptoms.
Our literature review on this topic yielded only one case of exertion‐associated cervical interspinous bursitis [11]. Perka et al. reported one adult patient experiencing interspinous bursitis of the cervical spine following a 3‐h tennis match. To the best knowledge of the authors, this case is the second case of exercise‐associated cervical interspinous bursitis, and the first in the pediatric setting.
To the best of our knowledge, this is the first known reported case of exertion‐associated cervical interspinous bursitis in a pediatric patient, treated successfully with nonoperative management. Our patient was asymptomatic on clinical review 3 months post his index presentation and had returned to his full activities of daily living. We propose that the butterfly stroke may have been the precipitating factor, resulting in lower cervical spine spinous process impingement.
This hypothesis of mechanical irritation is consistent with the original description of Baastrup in relation to lumbar interspinous bursitis. Higher order imaging in the form of MRI played a key role in making an accurate diagnosis while also demonstrating evidence of interval improvement at outpatient review. We believe this case report highlights the importance of interspinous bursitis in the differential diagnosis of atraumatic spine pain.
Author Contributions
Daniel Patrick McKenna: conceptualization, data curation, investigation, methodology, project administration, writing – original draft. Martin S. Davey: conceptualization, writing – review and editing.
Patrick O'Toole: conceptualization, project administration, supervision, writing – review and editing. Flavia Alberghina: conceptualization, methodology, project administration, supervision, writing – review and editing.
Funding
The authors have nothing to report.
Ethics Statement
The authors have nothing to report.
Consent
In accordance with the journal's consent policy, we confirm that written informed consent was obtained to publish this report from our patient's parent prior to submission for publication.
Conflicts of Interest
The authors declare no conflicts of interest.