Home Forums FUO

Hi everyone,

I desperately needs thoughts or ideas about this case. Very long and complex, I'll try to simplify.

74-year-old male, reasonably healthy, the usual diabetes/cholesterol/blood pressure, history coronary stent 12 years ago. History diverticulitis, history kidney stones, history prostate cancer, latter treated by external irradiation. About a month ago patient started with what he thought was diverticulitis, with abdominal pain and night sweats, although pain in right lower abdomen, he started antibiotics that he had at home. No help, after a week he came in to see me, ordered CT of abdomen, showed large right ureteral stone. Stent was placed between kidney and bladder, after about five days lithotripsy done, after two weeks stent pulled. Patient never felt back to normal during this, he had been on antibiotics, but there was no definite urinary infection. The only positive urine culture was Candida.

Stent pulled, antibiotics stopped, but after five days felt much worse, high fevers, shaking chills. Three out of three blood cultures positive for Staph epi, methicillin-resistant, however according to sensitivities, these are two different strains. One bottle has one pattern of sensitivity, the other two bottles have different, but identical to each other sensitivities.

Extensive workup for infection otherwise negative, CT chest/abdomen/pelvis, repeat blood cultures, urine culture all negative. Transesophageal echocardiogram negative. Sed rate 130(!!!!). Preliminary Lyme negative.

Infectious disease consultant feels extremely strongly that staph epi is not a pathogen without some artificial nidus of infection (artificial joint, venous catheter, heart valve, etc.). He is convinced that the staph epi is all skin contaminant. He is actually favoring malignancy or connective tissue disorder, rather than infection.

Patient was on IV Vanco for about four days, seemed to get better, he's been off this for two days, seems to be getting worse.

Thoughts?

Thanks so much.

Gene

Well, I am a pediatrician, and I am not an ID doc, and I agree with the logic somewhat on the Staph epi, but if you have two identical MRSA strains and he was getting better on Vanco, I don't think I would pull out the textbook. Seems better part of valor would be to continue Vanco and figure it out later.

I would think 3/3 blood cultures is significant, maybe one is contaminant and that is why 2 different strains.

Infection Nidus- like a stone and ureteral stent?

Antibiotics until a better explanation is found.
Check fancy procalcitonin level?

Greg

If the onset was as sudden as it seems, this is infection until proven otherwise. Response to Vanco supports that. Partially treated infections are not fun to figure out. Hard to imagine Lyme with a ESR of 130, but I don't see Lyme out here. See if the Vanco makes an effect on the ESRs after a few days.

A sed rate of 130 is usually infection, and infrequently malignancy. Vasculitis is most likely of the autoimmune diseases with very high sed rate, but sudden onset makes that unlikely.

Drug reactions on top of another problem can be confusing.

Consider SPEP, IVP to look for an artificial nidus, rheum and onc consults, and tertiary referral if family stressed.

Let us know how it develops

I might be more convinced about a contaminant in the blood culture if just one, but 3/3 looks pretty strong, unless the technique for obtaining has serious problems. I like the idea of continuing Vanco and rechecking the sed rate later and comparing results with clinical improvement.

I would agree with the ID assessment of the culture results. When you have different organisms in different bottles, it sounds like a contaminant. When both organisms are skin colonizers, that seems like a slam-dunk. Gene, I hope it wasn't you who didn't wash well before drawing the cultures.

On the other hand, I wouldn't ignore the apparent partial esponse to the antibiotics.

Keep in mind that CAT scans of the abdomen are not "all-seeing". They can certainly miss diverticulitis as well as a colon cancer with micro-perforation. Was there diverticulosis on the CT?

74 year old man with right lower quadrant pain, acting like he is infected with a partial response to antibiotics? I would do a colonoscopy, looking for one of the above two diagnoses.

I guess to this hammer, they all look like nails.

Hold the antibiotics and re-culture. Some guidelines for SBE recommend up to 6 cultures (MKSAP question).

Gene,

I recommend the medieval treatment to prevent further kidney stones--drink lots of Natty Boh and ride a wild horse.

Seriously, can't think of anything that has not been mentioned already.

Is he definitely free of kidney stones? Has he had an ultrasound of his gallbladder? Has he had a HIDA scan?

[quote=]74-year-old male, reasonably healthy, the usual diabetes/cholesterol/blood pressure, history coronary stent 12 years ago. History diverticulitis, history kidney stones, history prostate cancer, latter treated by external irradiation.[/quote]
Is it because I am a pediatrician that this doesn't seem reasonably healthy, lol.
There is nothing worse than when patients just start themselves on antibiotics...

Bert,

Welcome to my world!

There are certain patients with relapsing conditions (UTI, occasionally diverticulitis) for whom I will trust to start antibiotics when they need, then let me know after the fact. You can certainly argue this, but usually it keeps them out of the ER.

Gene

This just might be a good case for an autopsy.

I wanted to get that in before jimmie thought of it.

Hi everyone,

Thanks for all the replies. I will flesh this out a bit. I can't give all the details, I received 59 pages of faxes for the first 36 hours of hospitalization.

The patient is the father of one of my staff, a rather high level staff...

The flareup occurred while on their long-awaited family vacation in Nags Head (Outer Banks) NC, 8 hours away. Initial hospitalization was at Outer Banks Hospital, an 18-bed hospital. Blood cultures were drawn there, sounds like technique was good. Vanco was started there.

After 36 hours, he was transferred to Norfolk Va, (still 5 hours away from me)community hospital with med school ties, but no house staff involvement.

ID guy continued vanco until blood culture results available, then decided since no manmade nidus for infection, to stop the vanco. Off vanco X 2 days, this evening clinically worse, with shortness of breath and hypoxia. BNP very high (?? 7000) CXR reported nl, repeat echo pending.

GB with stones, present for years, no evidence of infection here. Initial kidney stone 8 mm (ouch), after lithotripsy 7 mm "fragment" migrated back up into renal pelvis. Pt has had diverticulitis in past, this was always LLQ, initial pain this episode was RLQ. Kidney stone was R, I doubt he actually had diverticulitis this time. Several urine cultures neg, one pos Candida. CT abdomen with "Mild perinephric stranding."


White counts have NOT been elevated, typically 6000, automated so no real diff.

Lyme now equivocal, initial test neg second "may be positive"

Sed rate very high 130 on 3 occasions.

Difficult to say how long this episode has lasted. Initial night sweats were about 1 month ago, but fever only started about 8 days ago.

With high sed rate, positive blood cultures (I'm not sold on all 3 being contaminant) and now ??CHF sx, I think the single most likely dx is endocarditis. His normal TEE would of course argue against this, echo will be repeated tomorrow.

Gallium nuclear scan pending, first phase to be done tomorrow.

Overall, though, I would have him back on vanco. I would have not stopped it in the first place!

And I feel really really helpless, getting bits and pieces as this plays out.

Gene

I know you said this in jest, I'm afraid it may come true.

Point well taken, probably not the time to throw the humor around when it comes to this thread. (referring to my earlier comment about the medieval treatment)
But with right lower quadrant pain does he still have an appendix?
Hang in there Gene, and hope you get it figured out.

Sorry to joke, Gene.

Normal WBC sure makes infection less likely, and malignancy/vasculitis/drug reaction more likely. A reasonably healthy nonpediatric patient ought to be able to mount a white count to a life threatening infection.

I was thinking you had a sequestered infection that you could keep under some control while you probed, imaged, and diagnostic tested every pertinent part until you found the nidus, which could be cultured. The impression that Vanco seems to control this is important even if eventually wrong, since keeping him alive comes before the theory.

Malignancy and autoimmune disease are not going to take him soon like infection could. I imagine all the autoimmune tests are cooking. Since he sounds fragile, I would continue the antibiotics until everything between the stem and stern are checked.

Then comes the magic... steroids.

Gene,

This is highly unlikely and I think Virginia is about as far south as this goes, but would Babesiosis even be a possibility in the Outer Banks?

Do you have additional blood culture results now that he has been off of antibiotics? I like your SBE idea; multiple positive (or negative) blood cultures would be helpful.

Gene,

Please update. Enquiring mind wants to know.

I am still holding out that this is endocarditis, and that the initial TEE didn't see vegetations because it was early. I am still interested in what came up from repeat cultures and TEE. However, if a repeat TEE was negative, the chance of SBE drops significantly:

From Echocardiography in infective endocarditis; A Evangelista and M T Gonzalez-Alujas (Heart (2004 June; 90(6): 614-617)): "A negative TEE has an important clinical impact on the diagnosis of endocarditis with a high negative predictive value ranging from 86?97%. In patients with native heart valves, a negative TEE virtually rules out the diagnosis of infective endocarditis. However, in the study of Sochowski and colleagues, five out of 65 patients (7.6%) with an initially negative TEE were finally diagnosed with endocarditis: in three of the patients a TEE performed 1?2 weeks after the initial examination showed the presence of vegetations."

If the empiric treatment is directed at endocarditis from a GU/GI source, then the initial treatment with vancomycin alone was inadequate.
As Gene said, let us know, maybe save a future patient.

Hi everyone,

Sorry for the delay getting back to post. I have still been trying to sort things out, and I was waiting until I could give a reasonably helpful update.

The patient has been back in this area for about 2 1/2 weeks. He came back on Sunday evening, I saw him in the office on Monday, Tuesday morning he had a PICC line put in, and Tuesday evening had his first dose of IV vancomycin. He started with 1 g Q 12 hours, his trough level was a little high, so for the last week he's been on 1250 mg Q 24 hours. Today was day 16 of vanco. He's also on PO rifampin, which gives excellent tissue penetration.

Clinically, he is clearly improving. Fevers are gone, he feels better, feeling stronger, and his appetite is improved.

But night sweats continue, and he continues to lose weight, probably more than would be expected from his lack of appetite.

His sed rate remained unchanged, very high at around 120. How ever, his C reactive protein had been up around five, it's now normal at 0.9.

The clue comes in his CBCs. The last few CBCs have shown white count of about 11,000, but with a preponderance of lymphocytes, a sizable percentage of atypical lymphs, and a few smudge cells. His hemoglobin has consistently been around nine. So CLL seems likely. And this would make diagnosing sepsis more difficult, since he was not able to mount the typical leukocytosis, due to marrow suppression.

Patient will be having blood drawn early next week, this will include flow cytometry.

So the best I can put all this together is as follows:

Patient has been having symptoms, specifically night sweats, from CLL for about two months.

He then coincidently suffered from a large kidney stone, which required instrumentation.

Somewhere, probably related to the kidney stone or instrumentation, he became infected with staph epi. He was immunosuppressed from the CLL. The worst of his symptoms came from this infection. Where was the infection? My guess is still endocarditis.

He has been improving on the antibiotics, so the remaining symptoms, including the elevated sed rate, are from the CLL.

I plan one month of Vanco, which I think should be sufficient. By that time we should have a likely diagnosis of CLL. If that is the case, he probably will require treatment, due to the weight loss and night sweats.

So that's where we stand, and I'm reasonably sure that it's right.

Gene

Gene,

Thanks for sharing this difficult case and I really like this clinical corner and case section. Glad he is doing better with your care!!!