Safe Administration of Electroconvulsive Therapy in a Patient with Catatonia and Neuropsychiatric Lupus co-morbid with Fragile X Syndrome

Dear Sir,

We report on a complex patient with catatonia refractory to benzodiazepines in the setting of neuropsychiatric lupus co-morbid with fragile X syndrome (FXS), who was successfully treated with electroconvulsive therapy (ECT).

A 21-year-old woman with a history of FXS, Ehlers Danlos syndrome, myotonic dystrophy, attention deficit hyperactivity disorder, unspecified anxiety disorder, and obsessive compulsive disorder was admitted with a change in mental status, including visual and auditory hallucinations and catatonic symptoms. Her initial Bush-Francis Catatonia Rating Scale (BFCRS) score was 12. On admission, she also was found to have neuropsychiatric symptoms of newly diagnosed systemic lupus erythematosus (SLE). SLE treatment included prednisone, methylprednisolone, hydroxychloroquine, cyclophosphamide, and rituximab. While there was some benefit in inflammatory laboratory values with disease modifying treatment, her catatonia continued to progress. Plasmapheresis was considered, but risks outweighed potential benefits once she developed a deep venous thrombosis.

Lorazepam IV was started to address her ongoing catatonia, initially at 2mg three times (day #1 of hospitalization). There was some initial benefit to lorazepam but, she required increasing lorazepam doses to achieve the same benefit. By day #18, lorazepam had been increased to 2mg every 2 hours due to worsening BFCRS. This dose was maintained for one week (days #18–24) with the following BFCRS scores: 16, 16, 16, 15, 19, 25, 24. further escalation of intravenous lorazepam dosing was not conducted due to the risk for propylene glycol toxicity.

Despite treatment for immunologic and psychiatric symptoms, her mental status deteriorated, she had more frequent hallucinations, depressed mood, sleep disturbances and higher BFCRS scores. Given these refractory symptoms, ECT consultation was initially initiated on day #6 and reconsidered on day #19. Because of the patient’s Ehlers Danlos syndrome, pre-treatment medical evaluation included a transthoracic echocardiogram which was negative for aortic root dilation.

Due to the severity of the catatonia, ECT dose titration was not conducted, and she was treated with bilateral brief pulse ECT (PW 0.5 milliseconds, frequency 40 Hz, duration 8 seconds, current 800 mA, dose 256 mC) with an initial motor seizure duration of 29 seconds and EEG seizure duration of 72 seconds. Due to the diagnosis of myotonic dystrophy, which carries the risk of hyperkalemia with succinylcholine, anesthesia consisted of methohexital (1.25 mg/kg) with rocuronium (0.6 mg/kg). Over the next nine days she received five ECT treatments. BFCRS scores, starting on the first day of ECT treatment and after each ECT session were as follows: 13, 5, 7, 8, 7. Clinical global impression (CGI) was 3 “minimally improved” following 2 treatments, 2 “much improved” after treatment #3, and 1 “very much improved” following treatment #4. A lorazepam taper was initiated after ECT #2 from a dose of 24 mg daily over 23 days to 6 mg daily by discharge, one week after the last ECT treatment.

Immediately before treatment her Montreal Cognitive Assessment (MoCA) score was zero. MoCA increased to 15 following treatment #4. One week after her final ECT, MoCA increased to 23 (with 1/5 on delayed recall); she scored 21 six weeks after treatment (0/5 on delayed recall), and 21 ten weeks after treatment (3/5 on delayed recall). Six months after ECT, MoCA was 24 (4/5 points on visuospatial/executive function, 4/6 on attention, 1/3 on language, and 5/6 on orientation). She remains well from a catatonia perspective in the six months since ECT and has not required further ECT. She remains on lorazepam 2mg three times daily and immunosuppressants.

The exact pathophysiology of catatonia is still unknown. Dysfunction and decreased activity at gamma-aminobutyric acid A and dopamine D2 receptors, increased activity at N-methyl-D-aspartate receptors have been implicated, as have anomalies in the circuits connecting basal ganglia structures to the cortex and thalamus. Case reports describe the successful use of ECT for catatonia in the context of SLE.¹ Although case reports document catatonia in FXS patients,^2,3 there is a lack of literature for guidance on the safety and tolerability of ECT in this population. One case report describes a catatonic patient with Fragile X premutation who had elevated FMR1 and mRNA levels along with a significant deficit of FMRP, who responded to ECT, but the outcome was not encouraging in that cognitive sequalae were evident over time.³ For that patient, the etiology of cognitive deterioration was unclear and likely multifactorial, possibly related to chronic catatonia, cognitive dulling from high dose benzodiazepines and antipsychotics as well as a large number of medical and psychiatric co-morbidities, which included a seizure disorder, obsessive compulsive disorder, bipolar disorder, psychosis and pervasive developmental disorder not otherwise specified. In addition, there was a birth history of hypoxia and intra-ventricular hemorrhage with damage to the frontal lobes, and developmental delays. Lastly, serial neurocognitive testing showed a steady decline in verbal and nonverbal intellectual quotient domains starting at age 20 years old, prior to ECT treatment.³ Thus it is challenging to assign the cognitive decline in that patient to ECT treatment, although, as no other reports were available to influence risk stratification, that case report caused the family of our patient to hesitate to consider ECT (see below).

Generally, the most commonly reported domain of cognitive consequences following ECT is loss of autobiographical memory, and a decline in processing speed during treatment.⁴ Cognitive functions typically improve up to 6 months after ECT.⁴ ECT was discussed on day #6 then again on day #19 with the patient’s healthcare proxy (due to severe catatonia, the patient was unable to provide consent), who was acutely aware that a drop in cognitive ability would change the trajectory of the patient’s life. As such, they were hesitant to move forward with ECT until all options were exhausted. After days of maximum dosing of lorazepam, ECT was recommended again (day #19). However, due to ongoing concerns of potential cognitive impact of ECT on a patient with a pre-existing vulnerability, and a lack of published FXS-specific recommendations or cases, ECT did not start until day #28. There was additional concern for potential propylene glycol poisoning, due to intravenous lorazepam, which precluded administration of higher benzodiazepine doses.

We performed serial MoCA examinations to track cognitive function during and after ECT course, all of which were stable, including a MOCA done 6 months after ECT completion. There was marked improvement in catatonic symptoms which was sustained, and cognitive functions continued to be intact up to several weeks after treatment. The degree of intellectual disability in patients with FXS has been correlated with shortages in the FMR protein, and females with the full FXS mutation usually have higher FMRP levels due to the an additional X chromosome, which provides mosaicism in FMR1 expression.⁵ As our patient was female, high functioning at baseline, with good cognitive reserve, this may partly explain why she tolerated the procedure, and why cognitive sequelae were not a significant outcome.

This case adds to the scarce literature on the relative safety and efficacy of ECT in this population and suggests that FXS should not be viewed as an absolute contraindication to ECT. Given the high mortality rate of catatonia, we would recommend ECT for patients with FXS non-responsive to medication treatment. Further research is needed and consideration remains necessary to ensure benefits are weighed effectively on a case by case basis.