AN ALBUM OF URINARY MICROSCOPY IMAGES IN A CLINICAL CONTEXT
G.B. Fogazzi, Milan, Italy
S. Verdesca, Milan, Italy
The next case deals with a woman with a long history of microscopic haematuria of unknown origin.
Isolated microscopic haematuria is a condition characterized by an “abnormal” number of red blood cells in the urine (even though there is no agreement at all about the number of red blood cells to considered as “abnormal”) in the absence of proteinuria. For some authors the definition should also include the presence of a normal renal function and the absence of arterial hypertension.
In a recent paper published in prestigious “The New England Journal of Medicine” it was stated that the single most important test in the evaluation of a patient with haematuria is the microscopical analysis of urine. This because “it often distinguishes glomerular from non glomerular bleeding.”
Therefore, for patients with isolated microscopic haematuria we propose the following algorithm:
in which we put the urine sediment examination in the very first steps of the diagnostic work-up, just after the clinical history and the physical examination, in the same position with the measurement of serum creatinine and of urine protein excretion. This because we believe it is important to know, from the very beginning of the clinical evaluation, whether the haematuria is glomerular or non glomerular.
What do the words “glomerular” and “non glomerular haematuria actually mean? The story starts in 1982 when Fairley and Birch published in “Kidney International” the seminal paper shown in the slide.
In that paper they demonstrated that two main types of erythrocytes can be found in the urine. One type is represented by the so-called “glomerular or dysmorphic” erythrocytes, which are characterized by irregular shapes and appearances, and differ remarkably from the red cells we find in the peripheral blood.
The other type is represented by the so-called “isomorphic or non glomerular” erythrocytes, which are identical with the red cells we find in the peripheral blood.
After the paper by Fairley and Birch, many other papers were published in the 1980s by different investigators, all of which confirmed the utility of the examination of urinary erythrocyte morphology as a tool to identify the glomerular or non glomerular origin of the haematuria. However, as time passed it became evident that this examination requires experience, it is exposed to the risks of low inter-observer reproducibility, and is limited by the fact that very different criteria were used by different authors (for instance, for Fasset et al , a haematuria was glomerular if > 80% erythrocytes were dysmorphic; for Tomita et al,  if dysmorphic red cells were >15%, for Loh et al  if >3 morphological subtypes of erythrocytes were present, etc).
Luckily enough, a paper, published by Hans Köhler and co-workers in “Kidney International” in 1991, simplified this rather confused situation by identifying a unique subtype of dysmorphic erythrocyte, the so-called “acanthocyte”, which could be used as a reliable indicator of glomerular bleeding. This finding was confirmed by other authors, and today we know that the presence in the sample of ³5% acanthocytes indicates a glomerular bleeding with a 56%-100% sensitivity and 97%-100% specificity.
This is a classical example of an acanthocyte as seen by scanning electron microscopy (SEM),
while these are acanthocytes as seen by conventional phase contrast microscopy (arrows). As you can easily see, acanthocytes are nothing but erythrocytes with the shape of a ring from which on ore more “blebs” protrude.
With this background knowledge, microscopic haematuria of unknown origin should represent the ideal clinical situation for the evaluation of urinary erythrocyte morphology. In spite of this, only few papers have so far been published on the utility of this test in this clinical condition.
One is the paper published by Schramek and co-workers in 1991 in prestigious “Lancet”. The authors reported on 316 patients with isolated microscopic haematuria of unknown origin, who had been divided in 2 groups on the basis of the morphology of urinary erythrocytes. The 123 patients with a non-glomerular haematuria were submitted to a complete and immediate urological work-up: a urological disease was found in 85% of cases, while for the other 15% of patients no causes for the haematuria could be found. For the 193 patients with a glomerular haematuria (which was diagnosed when 100% of urinary erythrocytes were found to be dysmorphic), only a basic nephrological investigation was carried out, which did not include renal biopsy. At a mean follow-up of 42 months, 112 patients continued to have glomerular haematuria, while 5 underwent a complete and spontaneous reversion. In 2 other patients, glomerular haematuria transformed into a mixed haematuria, and in both of them a cancer of the urinary tract was found. Three other patients developed proteinuria, which suggested a progression of the renal disease. In our opinion, this paper demonstrates that the examination of urinary erythrocyte morphology is useful in the work-up of patients with microscopic haematuria, especially for the identification of urological disorders.
The other paper, which was published by McGregor and co-workers in 1998 in “Clinical Nephrology”, explored the aspect which had not been investigated in the previous paper. Seventy-five adult patients with microscopic haematuria of unknown origin were all submitted to renal biopsy. A glomerulopathy was found in 31 of 33 patients with a glomerular haematuria, and also in 31 patients of 42 with non a glomerular haematuria.
We also have some results in this field, which are going to be published in “Pediatric Nephrology”. We have submitted to renal biopsy 16 patients (10 children and 6 adults) all with isolated microscopic haematuria, which had been classified as glomerular after repeated urinary sediment examinations over time (2 to 8 per patients, for a total of 55 urine sediments)*. To define the haematuria as glomerular, we used the following criteria: the presence of ³40% dysmorphic erythrocytes and/or the presence of ³5% of acanthocytes and/or the presence of ³ 1 erythrocyte cast on 50 low power (160x) microscopic fields.
*We don’t usually perform renal biopsy on patients with isolated microscopic haematuria. These patients represented a selected group, in which the renal biopsy had been performed for various clinical reasons. We did not include in our study patients with non glomerular haematuria, because we considered renal biopsy unethical for such patients.
As you can see in the slide, we found a glomerular disease in 14 out of 16 patients, which in most instances was due to thin basement membrane disease or Alport syndrome. Thus, in our hands the finding of a glomerular haematuria was a good indicator for the presence of a glomerular disease. In addition, one of the two patients without glomerular changes at renal biopsy (in spite of a full investigation with light, electron, and immunofluorescence microscopy) we found a cluster of erythrocytes in the lumen of a renal tubule, which indicates the renal origin of the haematuria.
Due to the fact that we didn’t include in the study patients with non-glomerular haematuria, we can just give the sensitivity and a positive predictive value of our urinary diagnostic approach. If we consider the 55 urine samples, we have the figures shown in the slide: you see that the highest sensitivity is for ³5% acanthocytes while the lowest is for red blood cell casts, which however have the highest positive predictive value. If we consider the 16 patients, you see that we have very good sensitivity for both ³40% dysmorphic erythrocytes and ³5% acanthocytes, while this is very low for red blood cell casts.
Thus, even though the available studies are very few, are based on different methodologies, and have a number of other limitations, we think we can say that the evaluation of urinary erythrocyte morphology is useful to address the early diagnostic work-up in patients with isolated microscopic haematuria of unknown origin.
Now the clinical case.
In January 2007, a 66-year-old woman is seen in the clinic of our unit for persistent microscopic haematuria, which was first diagnosed in 1990, 17 years ago. Since then, the patient has always been looked after by urologists.
As you can see, the patient was submitted to an intensive diagnostic work-up over time: 8 ultrasounds of the kidneys and bladder; 1 CT scan of the abdomen; 1 intravenous pyelography; 3 cystoscopies, the second of which showed that were some areas of hypervascularisation and oedema at the level of the bladder trigone. This led to electrocoagulation of the hypervascular zone and to a third cystoscopy, which demonstrated the complete ablation of trigone abnormalities… while the microscopic haematuria persited unchanged. The work-up also included 8 urinary cytology examinations, which were all negative for cancer cells, 1 search for urinary tuberculosis, and 15 urine cultures. However, in 17 years the patient had never been offered the evaluation of urinary erythrocyte morphology!!
Thus, when at last the patient was sent to us, we examined 4 urine sediments over a period of 1 month, and found that all 4 samples suggested the presence of a glomerular haematuria. In the 1st sample, we had ³40% dysmorphic erythrocytes + 6% acanthocytes + 1 erythrocytic cast. In the 2nd sample we had 35% acanthocytes, which is 7 times higher than the cut off level. In the 3rd sample we found 10% acanthocytes. In the 4th, we found 5% acanthocytes and erythrocytic casts.
Thus, all samples examined showed a haematuria of glomerular origin. Serum creatinine, creatinine clearance, C3-C4, and IgA serum levels were all normal, and the search of autoantibodies was negative. Therefore, we explained to the patient that she had a glomerular disease (most likely, thin basement membrane disease or IgA nephropathy) which in 17 years has caused only isolated microscopic haematuria, and that she needed only periodical checks (twice/year) of serum creatinine, urinary protein excretion, and blood pressure.
I close the presentation of this case with the following question mark: how many invasive procedures could have been avoided if the urine sediment had carefully been examined at the beginning of the diagnostic work-up and not 17 years after the first diagnosis of microscopic haematuria?
THE FOURTH PART WILL BE PUBLISHED ON APRIL 30TH, 2008