Parkinsons: Accident prone or early Parkinson’s?

When Anel B’s* father received a Parkinson’s disease diagnosis, it provided some relief because it explained why he had been having so many falls.

She realised that it wasn’t just that her 82-year-old father was becoming accident-prone but, because his brain signals were not reaching his limbs, he was not able to put out a hand to stop himself falling. Earlier detection, she says, would have saved the family a lot of trauma, helped to plan the way forward, and her father would have received medication earlier.

Their ordeal started in 2018, when Anel’s father, Henry C, became disorientated at the bank. Soon after, he fell off a ladder while working at his coffee stall and when he came to, could not remember what happened. After that he had several falls, said Anel, twice through a window. “A neurologist did several tests, concluding that he had had a series of mild strokes. Blood thinners and a statin were prescribed.”

Still Henry’s condition deteriorated; he had more falls, and lost the ability to write and drive his car. The deterioration was faster than normal ageing, Anel noticed, and it seemed more serious than a mild stroke.

“Covid lockdown really exacerbated things. He would not stay inside — isolation was hard for him — but would go out without a mask on and not sanitise when he came back. I was worried he and my mother would become infected.”

Anel was getting daily calls from her mother saying she could not deal with the situation, and so she and her husband hurriedly built a cottage so her parents could be nearer. That was when she saw the way Henry’s health was declining. They consulted the neurologist again.

“After several tests, she told me what he didn’t have: his heart was fine, he had had no more strokes. She said it looked like Parkinson’s but she could not be sure. But by then our medical aid funds had run out.”

With the help of a psychiatrist and her occupational therapist daughter’s contacts in neuro-occupational therapy, Anel finally received a diagnosis to say her father’s condition “presented like Parkinson’s”. The psychiatrist prescribed Carbilev, which raises the dopamine levels Parkinson’s disease patients have a deficiency of, as well as other medication for the condition.

As soon as she could, early this year, Anel took her father to see a new neurologist, who confirmed three neurological conditions including Parkinson’s disease. She also located a pinched nerve in Henry’s neck, which required neurosurgery, and provided a letter for the medical aid so that he could get his medication on chronic benefits.

“If we had found out earlier it would have been easier for us as a family,” says Anel. “We had no clue what was going on. We couldn’t understand when he started saying weird things, and the falls. It was traumatic.”

It is not certain that earlier detection helps right now, as we don’t have any disease-modifying therapies that can halt or slow its progression.

Because of Henry’s rapid deterioration, he requires a full-time carer, which the family are struggling to pay for. “Now my father is on the medication he is not falling as much, but he is much weaker and is also sedated.”

Work being done with artificial intelligence (AI) offers new hope for earlier detection of Parkinson’s disease, the fastest-growing neurological disease in the world, which is often only picked up in the later stages.

In the study, AI was used to track the progression of the disease from breathing signals. It was undertaken by researchers at MIT, Rutgers University, the University of Rochester Medical Centre, the Mayo Clinic, Massachusetts General Hospital and the Boston University College of Health and Rehabilition. Called “Artificial intelligence-enabled detection and assessment of Parkinson’s disease using nocturnal breathing signals”, and led by Yuzhe Yang, it was published recently in Nature Medicine.

The model was evaluated among 7,671 individuals using data from several US hospitals and public data sets. 

The researchers say that thus far, no drugs can reverse or stop the progression of Parkinson’s disease. A difficulty in drug development and disease management has been the lack of effective diagnostic biomarkers, or characteristics. The disease is usually diagnosed based on symptoms that are related to motor functions such as tremor and rigidity.

However, the study says, motor symptoms tend to appear several years after the onset of the disease, leading to late diagnosis. “Thus, there is a strong need for new diagnostic biomarkers, particularly ones that can detect the disease at an early stage.”

It is also difficult to track the severity of the disease over time, and assessment of Parkinson’s disease progression relies on patient self-reporting or qualitative rating by a specialist physician.

In a way the groundbreaking study is building on work that was previously done, because a relationship between Parkinson’s disease and breathing was noted as early as 1817, in the work of James Parkinson, the English surgeon whose “Essay on the Shaking Palsy” first described the symptoms of what was later called Parkinson’s disease.

The study says the link between Parkinson’s disease and breathing was strengthened in later work, which reported degeneration in areas in the brainstem that control breathing, weakness of respiratory muscle function, and sleep breathing disorders. “Further, these respiratory symptoms often manifest years before clinical motor symptoms, which indicates that the breathing attributes could be promising for risk assessment before clinical diagnosis.”

The new AI-based system assesses night-time breathing to detect Parkinson’s disease, predict disease severity and track disease progression over time. The patient wears a breathing belt on their chest or abdomen, but the breathing signals can also be collected by transmitting a low-power radio signal and analysing its reflections off the person’s body.

The study says an important component of the design of the new model is that it “learns the auxiliary task of predicting the person’s quantitative electroencephalogram (qEEG) from nocturnal breathing… our system aims to deliver a diagnostic and progression digital biomarker that is objective, nonobtrusive, low-cost and can be measured repeatedly in the patient’s home”.

Though the new AI system is still far from being introduced in SA, Johannesburg neurologist Marcelle Smith believes AI will become more important in the diagnosis and management of all medical conditions, including Parkinson’s disease. “Although Parkinson’s disease is primarily a clinical diagnosis made on examination, AI might help in underresourced areas where there aren’t specialists with the appropriate skill to make the diagnosis.”

However, she said it was not certain that earlier detection of Parkinson’s disease would help. “To be honest, early detection is not that important right now as we don’t have any disease-modifying therapies that can halt or slow disease progression. This will hopefully change soon. Right now, early detection in the absence of problematic symptoms would only lead to a longer emotional and psychological burden of having such an illness.”

Early signs can also include nonmotor symptoms, including “loss of smell, constipation, depression and anxiety and REM sleep disturbances (acting out one’s dreams). These may occur even decades before the first motor symptoms, which may be tremor, slowness of movement and muscle stiffness”.

Smith says the cornerstone of treatment is levodopa, to replace the dopamine that the brain becomes deficient of in Parkinson’s disease.

“It is by far the most effective treatment but has the most side effects as well. There are additional treatments that can be used such as dopamine-agonists, amantadine and monoamine oxidase inhibitors. Each person requires a different treatment or combination of treatments to improve their symptoms. In certain cases, surgery or lesioning with gamma knife or ultrasound can be used.”

The US study says the findings provide evidence that AI can identify people who have Parkinson’s disease from their nocturnal breathing and can accurately assess their disease severity and progression.

“Importantly, we were able to validate our findings in an independent external Parkinson’s disease cohort. The results show the potential of a new digital biomarker for Parkinson’s disease. This biomarker has several desirable properties. It operates both as a diagnostic and a progression biomarker. It is objective and does not suffer from the subjectivity of either patient or clinician. It is noninvasive and easy to measure in the person’s own home. Further, by using wireless signals to monitor breathing, the measurements can be collected every night in a touchless manner.”

The research was partly funded by the Michael J Fox Foundation, and the former comedic actor is most publicly associated with Parkinson’s disease. Like many people who have gone through a process of transformation after diagnosis with a serious illness, Fox told Forbes magazine that his diagnosis of Parkinson’s was “a gift …  that keeps on taking, but a gift”.

What he meant was that his foundation has been able to spread awareness about Parkinson’s disease and advance research and patient engagement. Forbes said: “Mr Fox knew at the onset of his disease that he had the ability to change the world, by changing the way science plays a role in the world and the way patients play a role in science.”

His contribution to this study could have far-reaching effects, and help others to avoid experiencing what Anel’s family have been through.

* Not their real names