Case report

This 75-year-old retired prison officer initially presented with a mass in the tail of the parotid gland and hypoglossal nerve palsy. His medical history included type 2 diabetes and surgery 6 months earlier for nerve root impingement from cervical vertebral osteophytes. Biopsies confirmed squamous cell carcinoma. It was treated with parotidectomy, neck dissection and postoperative radiotherapy. Neither the primary site nor the cause of the hypoglossal nerve palsy could be established.

A year later, he developed neck swelling and recurrent laryngeal nerve palsy. Imaging revealed local recurrence below the previous radiotherapy field, which was treated with further surgery and radiotherapy. Positron emission tomography (PET) also detected a possible prostate tumour, but after investigation this was monitored via prostate-specific antigen assays.

Two years after his original presentation, a follow-up PET revealed another new concurrent primary tumour in the lower lobe of the right lung, which was surgically removed. The following year he developed multifocal neurology and an unsteady gait. MRI revealed widespread leptomeningeal metastases. He was too frail for further anticancer treatment and was referred to our palliative care service.

His main symptoms were neuropathic pain and spasm, treated with pregabalin, and tenacious stringy secretions. The latter were hard to expectorate and made swallowing food difficult. He was commenced on effervescent acetylcysteine 600 mg tablets once daily. As there was no improvement after 2 weeks, he was switched to carbocisteine oral solution 750 mg three times per day and saline nebulisers as required.

Ten days later, these conventional treatments had not helped, so we explained the reported use of β-blockade for similar secretions in MND. He was normotensive and felt the secretions were a significant burden, so we commenced atenolol 25 mg (5 mL) sugar free oral solution once daily. Atenolol was chosen because an oral solution was needed, and undesirable central effects are less common compared with the lipophilic β-blockers, propranolol and metoprolol.

A week later, he reported that the secretions were less stringy, more manageable and easier to expectorate. But his worsening mobility was making managing at home increasingly challenging, despite his stoicism, and he was, therefore, admitted to our hospice. During this 1-week admission, the atenolol was increased to 50 mg daily. His pulse and blood pressure remained normal. He reported further benefit from the increase, particularly that the secretions were less stringy.

Back home, where his condition gradually worsened, his family doctor tried reducing the atenolol as part of deprescribing, but the secretions became harder to expectorate and it was increased back to 50 mg daily. He experienced increasingly severe paroxysms of neuropathic pain that were treated with oxcarbazepine and methadone. A month later at home when he was unable to swallow, parenteral atenolol was considered but as he was too weak to expectorate, secretions were treated with hyoscine butylbromide and neuropathic pain with methadone and clonidine.

In 3.5 years, this male had experienced 3 separate primary tumours, 3 operations, 2 courses of radiotherapy and multiple investigations. Leptomeningeal metastases caused pain and weakness in his last months of life. But the secretions also caused significant distress. During the 2 months that atenolol was used, it did not remove the secretions but made them less tenacious and stringy, allowing him to cough and swallow more easily and making a useful contribution to his comfort. He bore this enormous symptom burden with great dignity and kindly consented to us sharing his experience in case it might help others.

Literature review and discussion

We searched PubMed, clinicaltrials.gov and WHO clinical trials database on 23 November 2021 using (((beta blockers) OR (*olol)) AND ((salivary secretion*) OR (sialorrhea) OR (sialorrhoea))). We looked for reports of bothersome secretions (e.g. cancer and neurological disorders) of all ages, languages and study designs. We excluded reports of β-blocker therapy for other symptoms. One-hundred and twenty-eight records were identified but none met the inclusion criteria and were rejected at title screening. One case series (n=16) was found via additional searching of references in review articles and guidelines.2

While no RCTs investigating β-blockers for thick secretions have been conducted, to the best of our knowledge, one case series found 75% of people with tenacious secretions due to MND benefited from β-blockers.2 A postal survey of the clinical practice of neurologists in managing salivary problems in patients with MND reported that 13% clinicians used β-blockers as second-line treatment for sialorrhoea and 7% used β-blockers for thick secretions specifically.1 To the best of our knowledge, no other data exist to quantify their effectiveness, either in absolute terms or compared with other therapies.

Although never formally studied, alteration of salivary secretion composition is the most likely mechanism of action leading to symptomatic benefit. The autonomic control of salivary secretions is broadly dichotomous: one group of cells within salivary glands, under the control of muscarinic and α-adrenergic receptors, produces a thin electrolyte-rich serous secretion on agonist binding, while a second group, under β-adrenergic control, produces thick secretions containing amylase and mucin when these receptors are activated. Studies in healthy volunteers have shown that 1-week treatment with atenolol (but not placebo or propanolol) significantly reduced salivary total protein content, while both atenolol and propranolol reduced salivary amylase concentration. Furthermore, these studies found that while salivary content was altered, overall salivary flow rate remained unchanged.3 Thus, we propose that:

• β₁-adrenoreceptors represent a potential target for treating tenacious secretions.

• The improvement seen is due to reduced salivary mucin secretion, rather than reduced overall volume of saliva.

While tracheal secretions may have also contributed to the patient’s symptoms, we do not believe these were affected by atenolol. Studies of human tissue suggest that these secretions are under the control of β₂-receptors, so atenolol—a β₁-specific antagonist—is unlikely to have affected their output. Furthermore, in vitro, atenolol was reported to not inhibit the secretory response of human tracheal epithelium to β-agonists (isoproterenol and epinephrine) when compared with a non-selective β-antagonists (propranolol) and a β₂-selective antagonist (ICI118551).4 Salivary—but not tracheal—secretion is, therefore, the most probable target.

Aside from the potentially direct pharmacodynamic advantage it may pose over non-selective β-blockers, atenolol might have a favourable adverse effect profile in patients with malignant disease because it penetrates the blood–brain barrier less readily than lipophilic β-blockers (e.g. propranolol and metoprolol). Indeed, in a crossover RCT, patients reported fewer neuropsychiatric symptoms such as nightmares and hallucinations compared with propranolol and metoprolol.5