NP & PA Student Blog

Smells as a Diagnostic Clue

Consider this post "part II" in response to faithful reader of our student blog, Mark Behar of Milwaukee, Wisconsin. His comment challenged me to follow up on the use of odors and smells to diagnose, or at least give us clues to how best to treat our patients' underlying pathologies. I can't refuse a good challenge. Thanks, Mark, because you made me do a little research that no doubt will help me become a more excellent PA. 

First, to help our learning, a brief case study:

HPI:

An 8-day-old white male infant presents to your ED with a 4-day history of lethargy and vomiting and the mother complains, "He smells bad even after I bathe him and he won't eat." The infant was full term and weighed 6 lb. 5 oz. At the visit today the infant weighs 5 lb 12 oz. The mother states he has had no wet diapers for 2 days. He is lethargic and vomits twice while in the ER. He will not take a bottle. 

Family and medical history are unremarkable. Child lives at home with two parents and healthy 2-year-old sister. 

PE:

On physical exam, the infant appeared dehydrated with sunken eyes, dry mucous membranes and no fat stores. The infant smells like sweaty feet or rotten cheese. 

Clinical Course:

After a full workup with lab tests, imaging and genetic testing, this patient is admitted to the NICU. You emphasized to the admitting NICU team that they should seek genetic testing as soon as possible because the smell of the child could indicate an underlying genetic disorder. The NICU team follows your suggestion, which helps expedite the patient's diagnosis: isovoleric academia (IVA), a rare autosomal recessive disorder that is pathognomonic for the "sweaty feet, cheese" smell. 

This patient has decreased production of the enzyme that breaks down leucine. Isovoleric acid builds up, resulting in its secretion, which leads to the odor on the skin and other effects of metabolic acidosis such as coma and eventual death. Is there any hope for this baby? Of course. The mother is instructed on a very specific low protein diet to avoid leucine as well as administration of carnitine and glycine, which research has shown will help metabolize the acid. 

After a week-long admission and follow-up with a geneticist and dietitian, the patient returns home. By 1 month, he has returned to his birth weight. This child's life will be extremely complicated by this disease, but misdiagnoses could have been fatal.

Bottom Line:

So, back to Mark's point: smells can save a life. Of course, it was actually all the labs and the genetic testing that proved the diagnosis but there's some support that following your nose as a practitioner may not be a bad idea.  

IVA is very rare. Most diseases pathognomonic with an odor are, but a few interesting ones I came across to complete our "smeducation":

• Nicotine addiction. (Just kidding, but we all know it's true.)
• PKU = Musty, locker room towels
• Maple syrup urine disease = (you'll never guess) maple syrup
• Hypermethioninemia = breath smells like cabbage
• Cystinuria = urine smells like rotten eggs
• Methionine malabsorption = malt or hop
• Cyanide poisoning = breath smells like almonds
• DKA = breath smells fruity
• Kidney failure = breath smells like urine
• Ileus = breath smells like solid human waste
  
• Tuberculosis = reported to smell like beer
• Malignant melanoma = smells like gasoline
  

Smelling down your patients is never a walk in the park to smell the roses because usually sick people don't smell like roses. For example, I came across a study that money could not pay me to participate in: smelling the elderly's incontinence pads to detect bacteriuria. It was only 33% accurate. I wonder if that was just the fault of the researcher's smell fatigue. I can't blame them. 

In medicine, a reality of our jobs is billing: we have to make money and prove stuff with our money-making tests. For this reason, diagnosing based on smell is tricky because our accuracy is rather pathetic if used alone without other costly tests.

Although we humans aren't always the most accurate sniffers, dogs, who have long been known to exceed humanity's abilities in a myriad of skills, prove to be super smellers in medical diagnosing. There's some other interesting research about dogs being able to rat-out a patient with schizophrenia from their smell. A recent Wall Street Journal piece highlighted the role of dogs in pediatric Type I diabetics to assist families in avoiding hypoglycemic emergencies.

These dogs don't come cheap and cost between $18-20K. I'm jealous of our furry cousins because if I could make that kind of money smelling, my student loan debt would be so much less daunting. Dogs are definitely where more researchers are putting their efforts, but don't let this dissuade you from smelling.

This brief post just scratches the surface of all that lies under the power of our noses. I suggest you do some follow-up research, especially with regards to "aroma therapy" for reminiscence. You never can tell what you'll smell along the way.

References

1. Kazushige et. al.  The scent of disease: volatile organic compounds of the human body related to disease and disorder.  Journal of Biochemistry Review: 2011;150(3):257-26.  Available from: http://jb.oxfordjournals.org/content/150/3/257.full.pdf+html
2. Midthun S.J., Paur R., Lindseth G.  Urinary Tract Infections: does the smell really tell?  Journal of Gerontologic Nursing.  2004 Jun;30(6):4-9.
3. Pavlou, AK.  Sniffing out the truth: clinical diagnosis using the electronic nose.  Clinical Chemistry Laboratory Medicine.  2000 Feb;38(2):99-112.  Available at: http://www.ncbi.nlm.nih.gov/pubmed/10834396
4. Rakheja et. al.  A neonate with hyperammonemia.  Lab Medicine.  2005 Jan; 36 (1).  Available at: http://labmed.ascpjournals.org/content/36/5/292.full.pdf
5. Linebaugh, K.  Service Dogs Pick up Scent of Diabetes Danger.  The Wall Street Journal.  2012 Dec 10.  Available from: http://online.wsj.com/article/SB1000142412788732400110457816342312970336.htm