Disseminated Lymphoplasmacytic Lymphoma to the Sympathetic Chain Causing VT Storm

Graphical Abstract

Ventricular tachycardia (VT) storm is typically defined as 3 or more episodes of sustained VT or ventricular fibrillation (VF) over a 24-hour period; this includes arrhythmias triggering implantable cardioverter-defibrillator (ICD) therapies in the form of antitachycardia pacing or ICD shocks.¹ VT storm is associated with high morbidity and mortality; thus, appropriate treatment is essential. Treatment options may include a combination of antiarrhythmic medication, radiofrequency ablation, deep sedation, or cardiac sympathetic blockade.¹ There are multiple mechanisms for VT in patients with cancer, including direct cancer cell myocardial infiltration, QT prolongation from cancer therapy, immune checkpoint inhibitor myocarditis, and electrolyte imbalances.

Case

We present a case of an unusual cause of VT storm in a patient undergoing treatment for disseminated lymphoplasmacytic lymphoma (LPL) that highlights the importance of a multidisciplinary approach to treatment involving oncology, electrophysiology, and cardiac surgery. The patient is a 68-year-old man with a past medical history relevant for complete heart block status post dual-chamber pacemaker (DC-PM), doxorubicin-induced cardiomyopathy, and relapsed stage IV LPL. He was first diagnosed with LPL 24 years prior and was stage IV at initial diagnosis with extranodal involvement of bone. His prior treatments included cytotoxic chemotherapy with 6 cycles of cyclophosphamide, doxorubicin, vincristine, and prednisone followed by 4 cycles of rituximab, an allogeneic stem cell transplant, proteasome inhibitors, and Bruton tyrosine kinase inhibitors including acalabrutinib and ibrutinib. He was diagnosed with cardiomyopathy (ejection fraction [EF] of 35%), 5 years before the current presentation. His cardiomyopathy was clinically stable and treated with valsartan 40 mg and dapagliflozin 10 mg daily; he was unable to tolerate additional agents because of baseline hypotension. Two and a half years before the current presentation, he developed intermittent complete heart block and underwent DC-PM placement. At the time of DC-PM, his left ventricular ejection fraction had improved to 47%, and a defibrillator was not indicated. Device interrogation 1 month before the current presentation demonstrated that he was pacing in the right ventricle 10% of the time.

The current presentation began during admission for initiation of venetoclax therapy to treat LPL disease progression. He was administered his first dose of venetoclax 200 mg without complication. After his second dose, he suffered a VT arrest; return of spontaneous circulation (ROSC) was achieved after 1 external shock and 2 minutes of chest compressions. His EF was measured after the arrest and had decreased to 24%. To investigate whether his VT was because of new ischemia, he underwent coronary angiography, which showed nonobstructive epicardial coronary artery disease. His serum potassium and magnesium levels were normal, and his QTc interval was 457 milliseconds with underlying ventricularly paced rhythm. Given baseline neutropenia from the patient’s LPL and chemotherapy, amiodarone was considered relatively contraindicated as there was concern this could worsen his neutropenia. Additionally, there is an established drug-drug interaction between amiodarone and venetoclax that raises venetoclax levels in the blood, increasing the risk of venetoclax toxicity. Because the patient was at high risk for recurrence of VT and his EF was now severely reduced, he underwent right ventricle lead extraction and upgrade to a biventricular implantable cardioverter-defibrillator for cardiac resynchronization therapy to facilitate resumption of venetoclax therapy and secondary prevention of sudden cardiac death. The patient was discharged on oral mexiletine for VT suppression.

The patient had not yet resumed venetoclax therapy when he was readmitted 3 months later for syncope with 16 ICD shocks for polymorphic ventricular tachycardia (PMVT), which were confirmed on device interrogation. The patient was admitted to the cardiac intensive care unit for management of VT storm. Laboratory data showed serum potassium 4 mmol/L and magnesium 1.7 mmol/L. His initial high-sensitivity troponin was 89 ng/L and peaked at 357 ng/L after 4 hours, which was thought to be secondary to multiple ICD shocks. The patient was started on intravenous lidocaine for VT suppression. Despite lidocaine therapy, the patient had refractory VT. Given that the VT was polymorphic, it was felt unlikely to be scar-mediated. VT ablation with Purkinje denetworking was considered, but without a stereotypic triggering premature ventricular contraction, this was felt to be of lower overall benefit. Instead, the patient was referred for left-sided stellate ganglion block (SGB) with 0.25% bupivacaine, which significantly reduced his VT burden.

Given the therapeutic response to sympatholysis, he was referred for definitive therapy with bilateral sympathectomy. Pathology from the left and right sympathetic chains demonstrated a dense infiltration of lymphocytes surrounding nerve ganglia that stained diffusely positive for the B-cell marker CD20 (Figure 1); these findings confirmed the patient’s known B-cell lymphoma infiltration into the sympathetic chains. Similarly, the patient had undergone previous lymph node and bone marrow biopsies throughout his cancer treatments, which also stained positive for CD20. After the sympathectomy, the patient resumed venetoclax therapy. Serial device interrogations postoperatively and during venetoclax treatment revealed no further episodes of VT. Whole-body positron emission tomography 9 months later showed no fluorodeoxyglucose-avid disease after 3 months of venetoclax 200 mg daily therapy. His EF gradually improved; 3 years after his index presentation, his EF was 50%, and he remained free of heart failure symptoms.

Figure 1.

Histopathlogy of the Sympathetic Chains

Arrows point to a dense lymphoid infiltrate surrounding nerves (A, B, E), ganglia (C, F), and perineural adipose tissue (D). The infiltrate is predominantly small, mature-appearing lymphocytes with occasional plasmacytoid morphology. Immunohistochemical staining showed that the infiltrate is diffusely positive for CD20 (E and F).

Discussion

The incidence of disseminated lymphoma to the sympathetic chain are rare. To our knowledge, there are few previous case reports. Prior reports did not describe VT but rather the diagnostic challenges associated with this rare disease.² In this case, we postulate that the mechanism of VT was enhanced sympathetic nervous system activation because of LPL infiltration of the sympathetic chain. The current evidence-based management of VT largely involves medications including antiarrhythmics, beta blockers, defibrillators, or cardioversion; catheter ablation is an additional treatment strategy for monomorphic VT. Sympatholysis from beta-blocker therapy can be helpful for VT suppression, particularly given the proposed mechanism of VT in this case. In order to achieve pharmacologic VT suppression, however, high doses of beta-blockade are typically required.³ In our patient’s case, beta-blockade was limited because of hypotension. Cardiac sympathetic denervation for treatment of refractory VT/VF, using either SGB or sympathectomy, is given a Class IIb recommendation for patients who are not candidates for or fail beta-blockade, antiarrhythmic drug therapy, or catheter ablation, as was the case with our patient.⁴

SGB is a temporizing measure used in patients with refractory VT. An additional advantage of SGB is that reduction in VT burden following the procedure can be used clinically to identify patients who would benefit from subsequent sympathectomy.^3,4 Local anesthetic, such as lidocaine or bupivacaine, is injected at the C6-C7 level. SGB is usually performed first on the left side, as it is postulated that the left stellate ganglion contributes more to sympathetic tone than the right.⁵ Bilateral SGB is typically pursued in intubated patients because of the risk of phrenic nerve paralysis. Unilateral or bilateral SGB typically suppresses ventricular arrhythmias for several days, although the timeline varies given different plasma dissipation times of the local anesthetic used.⁵ A recent multicenter registry evaluating the efficacy of SGB in 117 patients with refractory VT/VF showed significant reduction in the median episodes of VT/VF 24 hours before the procedure from 7.5 (Q1-Q3: 3-27.0) to 1.0 (Q1-Q3: 0.0-4.5) 24 hours postprocedure.⁶ Similarly, data demonstrating the benefits of sympathectomy are primarily from observational studies. For example, a study of VT recurrence after sympathectomy in 20 patients showed freedom from sustained VT or ICD shock after 1 and 4 years was 60% and 55%, respectively. Over a median follow-up of 1,267 days, 11 (55%) patients remained free of VT post sympathectomy. The median time for VT recurrence after sympathectomy was 90 days.⁷ Another study conducted in 121 patients with ICDs who underwent left-sided or bilateral sympathectomy showed that 1 year postsympathectomy, 58% of patients remained VT free and did not experience any ICD shocks.⁸ Overall, these data suggest that sympathectomy can be an effective treatment option for VT suppression in patients who fail conventional VT treatments. Some of the risks and long-term consequences of sympathectomy, however, include Horner syndrome, pneumothorax, hemothorax, and neuropathic pain.

Histologic studies of patients with VT demonstrate the important role of inflammatory cells in the generation of arrhythmias. In a study of patients with cardiomyopathy who underwent sympathectomy for refractory VT, the ganglia cells were characterized by leukocyte infiltration, the presence of CD3⁺ T cells, and myeloperoxidase-positive neutrophils in the parenchyma, which was not seen in organ donors with healthy hearts.⁹ In another study of isolated stellate ganglia from patients with long QT syndrome or catecholaminergic PMVT who had failed antiarrhythmic therapy and underwent stellectomy,¹⁰ the stellate ganglia demonstrated that concentrations of CD3+ and CD8+ T cells were significantly higher in the ganglia of long QT syndrome/catecholaminergic PMVT cases than in healthy control groups, suggesting that T-cell-mediated cytotoxicity toward ganglion cells may boost adrenergic activity.

Conclusions

Disseminated LPL to the sympathetic chain is rare, and its recognition as a potential cause of VT storm requires a high index of suspicion after common etiologies are ruled out. In this case, the patient did not have obstructive coronary artery disease, significant electrolyte abnormalities, or marked QTc prolongation to explain his PMVT. The patient had VT refractory to several antiarrhythmic therapies. Subsequently, he underwent bilateral sympathectomy, which demonstrated LPL infiltration of the sympathetic chain. We postulate that the mechanism of VT in this case was enhanced sympathetic nervous system signaling because of LPL infiltration. Sympathetic denervation is a reasonable option for incessant VT management in patients who have failed antiarrhythmic drug therapy and are not ablation candidates, particularly in patients with sympathetically mediated VT due to lymphoid infiltration as in this case.

Funding Support and Author Disclosures

Dr Upadhyay has served as a scientific advisor or speaker for Abbott, Biosense Webster, Biotronik, Boston Scientific, GE Medical, Medtronic, Philips BioTel, Rhythm Science, and Zoll Medical. Dr Smith reports a scientific advisor for Genentech/Roche, Genmab, AstraZeneca, and Foresight Diagnostics outside the submitted work. All other authors have reported that they have no relationships relevant to the contents of this paper to disclose.