Das AK, Pickett TM, Tungekar
GBM thickness -- a comparison of two methods of measurement in
patients with unexplained haematuria
Nephrol Dial Transplant
(Jul) 11:1256-1260 1996
Thin membrane nephropathy is a cause of micro- or macro- scopic
haematuria in adults which is associated with an abnormally thin basement
membrane. Mild proteinuria is common. Most cases are non-progressive but
a few develop nephrotic episodes, hypertension or chronic renal failure,
but there are no predictors of outcome. A small proportion of patients
have a family history. The diagnosis is considered when all other
causes of haematuria are excluded but diagnostic criteria are not fully
clarified perhaps because of the difficulties in obtaining normal controls
to establish the range of normal thicknesses related to age and sex.
In cases which undergo renal biopsy, an objective method must be used to
assess the thickness of the
glomerula basement membrane. This study compares two methods of measuring
1. The orthogonal intercept method as modified by Dische.
This requires 9 photos of the peripheral loops of 2 glomeruli at
a final magnification of 12000; then you measure the minimum distance
endothelial cell and epithelial cell membranes where the lines of an
overlaid 4 cm grid cut the basement membrane. You then calculate the
harmonic mean and correct for the obliqueness of cut.
2. A direct measurement method as described by McLay et al, NDT 1992; 7:191-
The same ultras were rephotographed at a final magnification of
approximately 30000 - the thinnest areas were chosen where epithelial and
endothelial cells were closest and did not seem to be tangentially cut.
The distance between the endothelial and epithelial plasma membranes was
measured, perpendicular to the GBM.
The advantage of the second method is speed: the authors estimate a
saving of 2.5
hours per case if two glomeruli are examined (the original article suggested
one was enough but this
was not thought to be sufficient by Das
Interestingly, the results of both methods were correlated, but the direct
measurement consistently produced results that were lower than the first
method, by an average of 80nm. They used different microscopes for the
different techniques so that may be one reason. The direct measurement
method also picks out the areas that look thin so you might expect lower
figures on that basis.
In a sense the absolute values are not so important as long as the method
has good repeatability. They calculated the coefficient of repeatability
by remeasuring cases by both methods and found that it was 39 nm for
method 1, 56 nm for method 2 on one glomerulus and 38 nm on two glomeruli.
95% of differences between repeat measurements would be expected to lie
within those values. So on that basis, the direct measurement method
based on two glomeruli would appear to be as good as the orthogonal
method. As well as time savings, because you actually select the areas by
the direct method, you don't have to worry about how to measure areas
where the epithelial cells have become disrupted for instance.
Having said that, those coefficients are not particularly good: the
authors suggest that you may improve repeatability by measuring more
glomeruli, at least in borderline cases. I think however that there must
be a point beyond which you will get disproportionately small amounts of
information relative to the amount of time it takes to accumulate the
The authors mention the difficulties in obtaining normal reference
values as these vary with age and sex. Some data has been published
(although as far as I know with no mention of the repeatability of the
measurements obtained) but clearly internal controls are better because of
variations in laboratory techniques and microscopes.
(Dr. Patrick Harnden, Leeds, United Kingdom)
The full text of this abstract is available from Oxford Press