Intracranial Hypotension Following Spinal Manipulation

Spinal manipulative therapies, including chiropractic and osteopathic maneuvers, are widely practiced for musculoskeletal complaints. However, serious complications such as cerebrospinal fluid (CSF) leak with subsequent intracranial hypotension (IH) have been described. The pathophysiological mechanism is presumed to involve mechanical stress on the spinal dura during high-velocity movements, leading to dural tears, particularly in the cervicothoracic region.

A team of Italian neuroscientists conducted a scoping review in accordance with the PRISMA extension for Scoping Reviews (PRISMA-ScR) guidelines, through a comprehensive search of PubMed and Scopus. They complemented the review with an illustrative case from their own institution.

The researchers identified 21 eligible papers, including 21 patients with IH following spinal manipulation. Most patients were women (81%), aged 29-54 years, and the majority underwent cervical maneuvers.

SMT techniques vary, most often involving high-velocity cervical maneuvers. The most frequent were axial tension with rotation in seven cases (33.3%), unspecified cervical manipulation in four cases (19%), and thoracic spinal manipulation in two cases (9.5%). Less common single-case techniques included rotation with hyperextension, combined cervical and thoracic mobilization, axial tension with lateral flexion, and occipital/shoulder tension technique (n = 1 case each).

Symptom onset was typically within the first week, and all presented with orthostatic headache, often accompanied by nausea, neck pain, tinnitus, or visual disturbances. Neuroimaging consistently revealed features of IH, with pachymeningeal enhancement and subdural collections as the most frequent findings; spinal imaging frequently demonstrated extradural CSF collections. Management was conservative in about one-third of cases, but most required epidural blood patching, which was effective in the majority. Surgical repair was necessary in rare, refractory cases, particularly in the presence of structural spinal abnormalities. Overall prognosis was favorable, with 95% of patients achieving full recovery.

The authors’ illustrative case highlights the potential for severe complications such as subdural hematomas and recurrence if the underlying leak is not addressed:

A 65-year-old patient without a previous history of headache presented with a progressively worsening headache, with orthostatic features, poorly responsive to medical therapy, that has lasted for the past 20 days. The patient denied any recent trauma. He reported having undergone cervical osteopathic manipulations within the past 3 months for recurrent cervicalgia. A brain MRI without contrast was performed, showing a large bilateral subdural hematoma with significant mass effect on the cortical gyri. The patient was admitted to the emergency department and underwent neurosurgical evacuation of a bilateral chronic subdural hematoma via burr holes. Subsequently, endovascular embolization of the middle meningeal arteries was performed as an adjunctive treatment to reduce the risk of recurrence. The surgical procedure was performed without complications. A cranial CT scan showed a reduction in the volume of the hematoma. Therefore, the patient was discharged. However, after a transient improvement in the symptoms, the patient continued to present a fluctuating headache without positional features, with four to five episodes per month. He was readmitted to our clinic and, upon arrival at the ER, a head CT scan showed an increase in pneumocephalus and a recurrence of the hematoma. The following day, an MRI of the neuraxis with contrast was performed, which revealed radiological findings suggestive of IH: pachimeningeal enhancement, subdural fluid collection, dural venous engorgement, cervical spinal longitudinal extradural collection, and effacement of the suprasellar cistern. The Bern score was 7. Given these findings, a surgical revision of the previous burr holes was performed without periprocedural complications. After the first day, a non-targeted epidural blood patch (EBP) was performed under local anesthesia by injecting 16 mL of autologous blood into the L3–L4 epidural space. The procedure was uneventful. A cranial CT scan showed satisfactory surgical outcomes, highlighting a reduction in the volume of the hematoma and of the pneumoencephalus. The patient was subsequently discharged with complete resolution of the headache.

The authors concluded that clinicians should recognize the possibility of CSF leaks after spinal manipulation, especially in patients with new-onset orthostatic headache.

I feel compelled to point out that, considering the multiple risks of upper spinal manipulations and the almost total lack of evidence of benefit from such treatments, the risk/benefit balance of spinal manipulation is clearly not positive. It follows, I think, that it would be wise for patients not to allow such therapies being carried out, and for healthcare professionals to discourage them.