The other bed was empty when I walked into the room, and my elderly patient’s family filled the space to form a semicircle of concern around her. I introduced myself and tried to talk to her. She answered mostly in gibberish. I wouldn’t be able to get a history from her.

A daughter explained that her mother had been falling for months, and for the last three days her cognition had grown steadily worse. Plus, she had been incontinent for the last two months. Her family doctor had prescribed medication for bladder control, but it wasn’t helping.

The patient denied everything her daughter said, but I didn’t think she was trying to deceive us. She had to be able to recall a truth in order to shield it in a lie. I switched to simpler questions.

“What’s your name?”

She told me.

“Where are we?”

A blank stare.

“Are we in a bank?”


“A church?”


“A hospital?”


That was good.

“What year is it?”


“That’s when you were born. How about now?”

Another blank stare.

She held my hand as she tried to rise from the chair. She could barely stand, and struggled to take a step, as if her feet were glued to the floor. I helped her to sit down.

We started by seeking an explanation for her rapid cognitive decline. Brain MRI was normal, meaning that there was no stroke. Metabolically she was fine, with no urinary tract or other infection, a common cause of acute cognitive problems in seniors. Given how quickly her cognitive impairment came on, it seemed a lot more like an encephalopathy — brain dysfunction from a virus or another external agent — than something chronic and intrinsic, but we had no satisfactory explanation for it.


Neurology recommended testing to look for causes of reversible dementia. Vitamin B12 deficiency, extremely low thyroid levels, and advanced syphilis are typical. The results of the studies were all normal.

We sent her for an MRI of the cervical spine. It showed some cervical stenosis (narrowing of the spinal canal), but it wasn’t terrible. The way her gait was impaired just didn’t look as if it was caused by cord compression, nor would her degree of stenosis explain the incontinence or cognitive impairment.

We kept digging and scrounged up CT scans and MRIs from previous years in her medical records. When we compared them with the new studies, we could see that the ventricles – the fluid-filled spaces in the center of the brain – were growing progressively larger over time.

That finding, plus her three main symptoms led us to suspect normal pressure hydrocephalus (NPH) – accumulation of cerebrospinal fluid enlarging the ventricles of the brain. Typically, NPH worsens over years rather than days, as our patient experienced.

The only thing left to do was a spinal tap to confirm NPH. We measure the pressure, then draw off about two tablespoons of cerebrospinal fluid and send it for about a dozen different lab studies. The main goal, though, was to see whether her symptoms improved once the fluid was removed.

There are a few ways to diagnose NPH. My preference is to test in the least invasive way possible, so my 80-something patient wouldn’t be subjected to general anesthesia and a procedure that might not help.

After her spinal tap, our patient went to rehab, then home. Her family reported that she improved dramatically over the following days. But about three weeks later, she was regressing, and came back for treatment.

Now that we knew that the spinal tap helped her so much, we were confident in recommending surgery under general anesthesia to implant a ventriculoperitoneal (VP) shunt. This device drains cerebrospinal fluid in a controlled way from the brain ventricle into the abdominal cavity, where it’s absorbed and excreted.

About 10% to 15% of all dementias are reversible, and one cause is NPH, though why some people develop it is not fully understood. The conventional wisdom is that the brain becomes more “floppy” with age and deforms under the normal pressure of cerebrospinal fluid in the ventricles. The way it deforms causes the classic clinical triad that we saw with this patient.

Surgery usually helps more with walking and incontinence than with cognition, but in our patient, I’m happy to report that the shunt helped all three. She’s living at home independently once again.

Patrick Connolly is a clinical associate of neurosurgery at Penn Medicine and neurosurgery chief, Virtua Memorial Hospital.