To the Editor: Finding effective biological treatments for posttraumatic stress disorder (PTSD) has been difficult (1). Case reports have suggested that stellate ganglion block, a procedure in which a local anesthetic is used to transiently block cervical and upper thoracic sympathetic ganglions, might be helpful (2, 3). Although stellate ganglion block is well characterized as a treatment for sympathetically mediated pain, the few reports have not used gold-standard psychometrics to define PTSD or measure improvement (2, 3). We report here on a series of patients who were treated with stellate ganglion block for PTSD and were monitored before and after the intervention using the Clinician-Administered PTSD Scale (CAPS), the gold standard for PTSD assessment.

Nine military patients with chronic PTSD were treated with stellate ganglion block. All of the patients had been treated for over a year in programs that used selective serotonin reuptake inhibitors, individual psychotherapy (cognitive processing therapy or prolonged exposure therapy), and other evidence-based treatments. They were in the process of being medically retired from the military, and they were all aware that stellate ganglion block was offered as an off-label indication and gave informed consent to the treatment. The patients were not required to stop any other treatment to receive the stellate ganglion block, and all chose to continue their medications. For stellate ganglion block, fluoroscopic and ultrasound guidance were used with a lateral or paratracheal approach. Ropivacaine, 7 mL of a 0.5% solution, was injected following negative aspiration and negative test dose. Injection was performed at the C6 level, which was confirmed by observing the position of contrast dye, but medication may have spread to the level(s) above and below. Minimal sedation guidelines were followed, and sterile technique was used. PTSD symptoms were assessed using the CAPS before the procedure and 1 week after stellate ganglion block. Outcomes were tracked after a single injection in seven patients, and a second treatment was performed 4 to 6 weeks later in two patients.

Transient neck pain and stiffness were universally present after the procedure, and most patients experienced transient Horner's syndrome. No serious event or complication occurred. After the stellate ganglion block, five of the nine patients (56%) experienced a clinically significant (>30%) reduction in PTSD symptoms, as measured on the CAPS 1 week after the procedure. Two patients experienced noticeable but not clinically significant (10%–30%) reduction in PTSD symptoms. The remaining two patients reported essentially no change in symptom reduction (<10%). Six of the nine patients reported perceived improvement in symptoms after the block. One individual who had been suicidal for the previous 2 years reported immediate relief. In patients who had regular follow-up, the benefit of the block on PTSD symptoms appeared to diminish within 1 to 2 months of the procedure, although symptoms did not always return to their previous level of severity. The results of two repeat treatments were also tracked. One patient who did not initially benefit had no improvement after a second block, whereas the individual with the greatest gains had symptoms remit after the second procedure.

This case series suggests that stellate ganglion block may be effective as a rapid treatment for PTSD. Unlike previous reports (2, 3), the use of the CAPS here allowed us to verify that the patients being treated had the chronic condition and to quantify improvement. Although a majority of patients experienced symptom reduction after the block, the benefit was not universal and appeared to be transient. Mulvaney et al. (3) previously reported longer-lasting improvements in two patients, and we are unsure why the benefits we observed did not last as long. Considering the treatment-resistant sample and the rapid nature of the response, we think it is unlikely that the benefits are attributable to placebo effect or to the other treatments the patients were already receiving. Randomized controlled trials should be conducted to test this effect.