I am a complete fraud; an imposter. I am not a neurosurgeon. I am not even a neurologist! I am a retired hepatologist who has spent more than 50 years on the faculty at Yale University. The only truly noteworthy entry in my curriculum vitae is that I am both a Professor of Medicine at Yale and Professor of Surgery (clinical) at the University of Miami. Until four years ago I had never even heard of normal pressure hydrocephalus, which I shall refer to as NPH. I learned about NPH the hard way—I contracted it.

MY STORY

In 1993 at age 67 I had retired. Early in my retirement I began to have trouble walking. My gait was slow and awkward, and gradually over 10 years it worsened until I was unable to walk at all. I lurched from place to place. The Yale neurologist that I selected to care for me thought I had a Parkinson’s disease-like syndrome caused by cerebral atrophy. My brain MR scan showed both dilated ventricles and sulci (fig 1). Even after I had developed urinary incontinence, the neurologist still failed to recognise the diagnosis of NPH. Three other senior neurologists diagnosed a variant of Parkinson’s disease. I was unaware of any dementia. I thought that my loss of short-term memory was due to “old age”. My wife noted in her diary that I was not as sharp, as quick or as funny as I had been. The Yale neurologist persevered in his diagnosis of cerebral atrophy. After five years he called my wife and me into his office and told us that the cerebral atrophy had worsened and would probably continue to do so. He predicted that I would lose intellectual function. He told us that there was no effective therapy, nor would there be. We thought that he was advising us to get our affairs in order. We cancelled our 50th anniversary celebration, informed our friends and relatives of impending disaster and planned to continue our life in Florida where I was a consultant for the Liver Transplantation Program at the University of Miami.

• Download figure
• Open in new tab
• Download powerpoint

Figure 1

T2-weighted MR scan showing my dilated lateral ventricles with enlarged cortical sulci.

By November 2002 I was 77 years old. I had to quit playing squash and bridge, my joie de vivre was disappearing, and my gait was rapidly deteriorating. Besides, double incontinence is a terrible way for a fastidious person to live. I could barely walk from the garage to my office at the University of Miami. I wrote from Florida to the Yale neurologist and asked him to authorise a MEDICARE-subsidised electric scooter for me. For reasons that I still don’t understand, he refused. I quit my job at the University and started to accumulate sedatives and analgesics in the hope that I could dispatch myself to a better world. My attending physician, an endocrinologist who is the Director of the Diabetes Institute at the University of Miami, couldn’t authorise the scooter (only neurologists, physical therapists and orthopedists can). However, he was so incensed by the neurologist’s refusal that he immediately referred me to a much younger neurologist who, he said, would approve the scooter and would also give me a “second opinion”.

Several days later the new neurologist approved the scooter and recognised that I had NPH. He told me all about this reversible neurological syndrome. Incidentally, the size of my ventricles did not appear to have increased over the five years of observation, despite rapid progression of the symptoms.

After a decade of depressing diagnoses I didn’t immediately accept that I had this relatively new, mysterious, reversible illness until my neurosurgeon had performed a lumbar puncture that induced an instantaneous, miraculous remission. I had had difficulty climbing onto the examining table, but 10 minutes after the drainage of 60 ml of spinal fluid had started I could walk again. It was such an immediate, dramatic change that I, as an objective scientist who does not believe in casual miracles, did not believe that it could have occurred so rapidly. But, as an objective observer, I assure you that this miracle had occurred. Immediately after the CSF drainage I could again walk normally and felt and acted like my old self. A ventriculo-peritoneal shunt (VPS) made the remission long lasting.

My abnormal gait, incontinence and demential symptoms all disappeared transiently after the drainage of CSF and persistently after the VPS.

A NEW RESEARCH CAREER

As an experienced academic physician with a lot of free time, I saw the opportunity to study this new, poorly understood, not fully accepted, reversible illness. At that time NPH was less than 40 years of age, and the total number of published articles about it was only about 700. I thought that I could quickly assimilate the whole literature of this unique syndrome. Neurosurgeons are very generous physicians. They welcomed a neophyte into the field. Dr Salamon Hakim, who is from Bogota, Colombia and who first described NPH in 1964,¹ and his three neurosurgeon sons, have practically adopted me into their family and their ongoing research programme. Many other physicians including Drs Raymond Adams, Robert Barr, C Miller Fisher, Michael Tuchman, Carsten Wikkelso and Linda Sternau have generously contributed to my education.

On learning that NPH is not a universally accepted disease, that a few experts still doubt its very existence and that the efficacy of ventriculosystemic shunts is questioned by some neurologists and neurosurgeons,² I promptly abandoned my interest in the liver and dedicated my new life to making NPH a familiar term for physicians and lay people alike. I consider myself to be an amateur “NPHologist”.

I am now four years into my new discipline. During this time I have encountered more than 20 patients who have or appear to have had previously undiagnosed NPH of which several have undergone successful shunt surgery. NPH has still not yet taken its rightful place in medicine. In America many physicians know little or nothing about it. Its pathogenesis is still not understood. The criteria for its diagnosis are still not definitively established. An often-heard practical diagnostic “criterion” is that the diagnosis of NPH has not been established until the patient has responded to a shunt. Clearly, we are able to treat NPH much better than we can explain it!

I have made some progress in my personal goals. I have appeared on TV in Florida and spoken on National Public Radio in Connecticut about NPH. I have written one previous article about NPH. A colleague and I have recently completed a survey of about 300 physicians entitled “What is known about NPH and when was it known?” (Conn, Lobo; unpublished data). I finally made the Big Time when I spoke at HYDROCEPHALUS 2006, but I felt intimidated because I thought that everyone in the audience knew more about NPH than I did.

COMPLICATIONS OF VENTRICULOPERITONEAL SHUNTS

Complications of ventriculoperitoneal shunts are an integral part of the NPH syndrome. I have experienced several.

Loss of hearing

About three months after insertion of my shunt I required adjustment of the valve because of a recurrence of incontinence and abnormal gait. My Medtronic Strata valve was extrinsically adjusted from performance level 1.5 to 0.5, which resulted in prompt clinical improvement and an overt loss of hearing. Audiometry confirmed a loss of 15–20 decibels across the whole Hertz spectrum. This occasional complication of ventricular shunts is usually transient,³ but in my case it appears to be permanent.

Subdural haematoma

The second complication appeared in January 2004, eight months after shunt insertion. A routine CT of my brain showed that I had large, bilateral, clinically silent, subdural haematomas that compressed both cerebral hemispheres (fig 2). The likely precipitating trauma, which I had considered trivial, had taken place in June 2003, seven months before, while I had been anticoagulated for atrial fibrillation. I had fallen in the bathtub. I had grossly underestimated the force involved in coup and contracoup. In fact, when asked if I had had any trauma, I answered, “No”. I had completely forgotten about it. To paraphrase an old joke: “It wasn’t the height from which I fell, but the sudden stop almost killed me.” Discontinuation of anticoagulation and re-adjustment of the shunt pressure gradient was followed by complete resolution within three months. The subdural haematoma was completely silent. There have been no recurrences.

• Download figure
• Open in new tab
• Download powerpoint

Figure 2

T2-weighted MRI showing large, asymptomatic, bilateral subdural haematomas with significant compression of both cerebral hemispheres. (A) Old blood and (B) new blood in the left subdural haematoma, and (C) old blood in the right subdural haematoma.

Retrograde, catheter-transmitted, bacterial infection of the central nervous system

Now I’d like to bring to your attention a new, hitherto undescribed complication of ventriculoperitoneal shunts. I term the syndrome “Retrograde, catheter-transmitted, bacterial infection of the central nervous system induced by coliform bacteria associated with diverticulosis, diverticulitis and peritonitis”. Its name describes its pathogenesis. Figure 3 shows diverticulosis including a diverticulum with a leak (arrow pointing towards the open end of the ventricular catheter which leads directly to the brain.) Several months after the ventriculoperitoneal shunt had been inserted, I developed pain and tenderness in the left lower quadrant of my abdomen, which increased over the next 48 hours. When rebound tenderness appeared I was admitted to hospital where a clinical diagnosis of acute diverticulitis was made. A CT scan of my abdomen with oral contrast medium showed diverticulosis of which I had not previously been aware, and diverticulitis with a microperforation (fig 4). When I saw the films I was alarmed to see how near the distal end of my peritoneal catheter was to the site of perforation of my bowel (7 or 8 cm). These catheters are intended to be one-way shunts that flow from the ventricles to the peritoneum, but I believe that sometimes shunts flow in the retrograde direction, especially when pressure in the peritoneal cavity increases, as it does in bacterial peritonitis. Furthermore, it is known that histological examination of the tracts from the infected peritoneal cavity to the ventricles show inflammatory cells and bacteria on the outside of the catheters, which indicates that such bacterial transpositions do not occur solely via their lumens. In my earlier incarnation as a hepatologist I had learned how devious and stubborn coliform bacteria can be. Indeed, in 1964 my colleagues and I described the spontaneous bacterial peritonitis syndrome,⁴ and predicted that bacterial peritonitis in patients with ventriculoperitoneal shunts could be transmitted directly to the brain. Although I have never seen such a case, I’m certain that among the thousands of shunts that have been inserted, bacteria must occasionally have managed that retrograde passage. It is a statistical certainly, based on some relevant facts about diverticulosis which are not common knowledge for most neurosurgeons. About three quarters of healthy people have diverticulosis by the time they are 75 years of age,⁵ which is the typical age of patients with NPH at the time of diagnosis. It is estimated that about 30% of them will develop diverticulitis at least once. Assuming that NPH does not confer some sort of immunity against diverticulosis, diverticulitis should on statistical grounds occur once, at least, in virtually every patient with NPH. Therefore, I believe that infections of the central nervous system have occurred, and I predict that they will continue to occur in shunted patients with NPH. Such coliform infections of the central nervous system have been reported after perforation of the bowel by peritoneal catheters.⁶ The pathogenesis is identical to that which I have hypothesised for diverticulitis. My own diverticulitis was effectively treated with antibiotics and as far as I know my brain was not infected.

• Download figure
• Open in new tab
• Download powerpoint

Figure 3

The relation between the end of the ventriculo-peritoneal shunt (VPS) in the peritoneal cavity and a diverticulum of the colon. The black dots represent colonic coliform bacteria.

• Download figure
• Open in new tab
• Download powerpoint

Figure 4

Axial CT scan with oral contrast material of my abdomen showing acute diverticulitis. The large white arrow shows the site of perforation. The other arrows show inflammation and gas bubbles.

CAN DIABETES BE PART OF THE NPH SYNDROME?

During 2002 my symptoms of NPH suddenly worsened. My gait abruptly deteriorated, I developed urinary incontinence and, shortly thereafter, faecal incontinence. In addition, I realised that I was no longer able to play bridge well, probably because of my failing short-term memory—I could not keep track of trump, an essential requirement to playing good bridge. Indeed, this may be considered to be the ultimate test of short-term memory because the number of trump is continually changing. In the midst of the appearance of all these diverse symptoms of NPH I was discovered to have a fasting blood sugar of 165 mg/dl, which is diagnostic of diabetes mellitus. Was this temporal relation chance? Why did all of the symptoms of NPH appear at about the same time as the diabetes?

The diagnosis of diabetes came as no surprise. Both of my parents and all three of my siblings had had type 2 diabetes. Every year or two since I was 25 years of age I had had a glucose tolerance test performed using the Conn glucose tolerance technique (100 g of glucose after three days on a 300 g carbohydrate preparatory diet).⁷ I had to use this specific method because J W Conn, the endocrinologist who devised that test, was my older brother. These glucose tolerance tests had always been normal, but the diabetes appeared at approximately the same time as the symptoms of NPH. I doubt that this was a coincidence. Rather, I propose that diabetes mellitus may be a symptom of NPH, just as abnormal gait, incontinence and dementia are.

An association between the hypothalamus and diabetes has long been known. In 1858 Claude Bernard showed,⁸ and Banting and Best confirmed, that ablation of part of the hypothalamus of dogs may promptly induce transient hyperglycaemia. Diabetes has been reported in patients with hypothalamic tumours and in hydrocephalic patients with acute ventricular obstruction and increased intracranial pressure. No such observations have yet been reported to my knowledge in patients with NPH. However, Jacobs, in 1977, reported that the prevalence of diabetes is four times greater in patients with NPH (52%) than in age-matched control subjects (12%).⁹ I am unaware of other patients in whom the classic manifestations of NPH are related to abnormal carbohydrate metabolism, but Jacobs’ finding that diabetes is so common in NPH suggests that they may well exist. Certainly, confirmation of the increased prevalence of diabetes in NPH, of the relation of carbohydrate metabolism with ventricular volume, with central nervous system pressure and with their reversibility by ventriculosystemic shunts are fruitful areas for research. Furthermore, cases of the syndrome of the inappropriate secretion of the antidiuretic hormone have been reported in patients with NPH.¹⁰ They have been postulated to be caused by compression of the hypothalamus by the dilated ventricles—the same mechanism for which I propose for diabetes in NPH. My type 2 diabetes improved after weight reduction and has remained easily manageable on oral agents (sitagliptin plus phenformin) for the past four years. The shunt pressure gradient has been adjusted several times for various reasons.

A PHILOSOPHICAL SUMMARY

By virtue of a prolonged, erroneous diagnosis in my case I was afforded the opportunity to observe the whole course of this slowly evolving illness from its insidious onset to its nearly terminal stages more than 10 years later. The clinical features, the dramatic diagnosis which had been made for irrelevant reasons, the striking remission induced by a shunt, and the strange behaviour of the attending neurologist had all been documented by my wife, who has kept a daily diary throughout our married life. My long experience as an academic hepatologist and a medical editor serendipitously prepared me to recognise a hitherto unknown complication of ventriculoperitoneal shunts, to sense the association of diabetes mellitus in some patients with NPH, in part from personal experience, and to deal with complex clinical concepts that have permitted the insight that the characteristic clinical triad of NPH—gait abnormalities, incontinence and dementia—may sometimes be a tetrad (or a quadrad) when diabetes or inappropriate secretion of antidiuretic hormone is present, a dyad (when either incontinence or dementia is absent) or conceivably even a monad (when only the gait abnormality is present).

AND FINALLY …

In closing I would like to make a personal statement. I consider myself to be extremely lucky. As an agnostic I have almost come to believe that there may be some Higher Authority with a flare for the dramatic who creates such bizarre scenarios as my case history. In fact, at times I almost feel that I am fortunate to have been afflicted with this unique disease.

AUTHOR’S NOTE

This article is based on a talk I gave at the International Hydrocephalus Congress in Gotebörg in September 2006.