Dear Editor:
The article by Hennessy and co-authors is important and interesting,
and the last three words of the abstract are essential (The utility of
relative afferent pupillary defect as a screening tool for glaucoma:
prospective examination of a large population-based study in a south
Indian population. BJO Online First, February 24, 2011, DOI: 10.1136/BJO.
2010.194217). Their conclusion is, "The authors find that APD assessed by
the swinging flashlight test is a poor screening tool for glaucoma in this
setting." The authors state, ". . . the low prevalence of afferent defect
. . . 60/5150 (1.2 percent in those referred for evaluation) led to a
large number of true negatives. It was previously reported by
Ramakrishnan and colleagues15 that "the prevalence of any glaucoma in the
same population was 2.6 percent." However, the authors only found 77
cases of glaucoma in their 5150 referred cases. Thus, the percentage of
patients found with a simple test performed by individuals who are not
expert came up with a percentage of 1.2 percent, and in that study
population the expert glaucoma specialist found a percentage of 1.5
percent with glaucoma, not the 2.6 percent they mention. Furthermore, it
is clear that the technicians screening for afferent pupillary defects
were not highly reliable in recognizing afferent pupillary defects. The
technicians found 11 afferent pupillary defects, but ". . . only one was
verified to have a pupil defect at the time of the comprehensive
examination. . ." My major concern is that clinicians around the world
will generalize unwisely from a comment made by this group of highly-
respected authors. Specifically, they state, "Many academic institutions.
. . still teach residents and fellows to consciously look for an APD as
part of any initial clinical examination. . ." The authors are clearly
suggesting that is inappropriate. However, there are few tests that are
as quick and as inexpensive, and whose results are so important. The
presence of an afferent pupillary defect is a sure sign of significant
disease. The clinical significance of its presence, then, is close to 100
percent. To test for an afferent pupillary defect properly requires that
the room be dark, that the light be bright, that the patient be looking in
the distance, and the light be held in each eye for three seconds. The
authors have verified themselves that clearly the testing was not well
done. The overwhelming majority of individuals seen by physicians have
borderline findings, in which it is not clear whether the involved person
is merely a variant of normal or has actual disease. To discard a test
that can quickly and inexpensively solve that common dilemma seems unwise.
Sincerely,
George Spaeth, M.D.
Louis J. Esposito Research Professor
Wills Eye Institute/Jefferson Medical College
Philadelphia, PA
Conflict of Interest:
None declared
Dear Editor:
The article by Hennessy and co-authors is important and interesting, and the last three words of the abstract are essential (The utility of relative afferent pupillary defect as a screening tool for glaucoma: prospective examination of a large population-based study in a south Indian population. BJO Online First, February 24, 2011, DOI: 10.1136/BJO. 2010.194217). Their conclusion is, "The authors find that APD assessed by the swinging flashlight test is a poor screening tool for glaucoma in this setting." The authors state, ". . . the low prevalence of afferent defect . . . 60/5150 (1.2 percent in those referred for evaluation) led to a large number of true negatives. It was previously reported by Ramakrishnan and colleagues15 that "the prevalence of any glaucoma in the same population was 2.6 percent." However, the authors only found 77 cases of glaucoma in their 5150 referred cases. Thus, the percentage of patients found with a simple test performed by individuals who are not expert came up with a percentage of 1.2 percent, and in that study population the expert glaucoma specialist found a percentage of 1.5 percent with glaucoma, not the 2.6 percent they mention. Furthermore, it is clear that the technicians screening for afferent pupillary defects were not highly reliable in recognizing afferent pupillary defects. The technicians found 11 afferent pupillary defects, but ". . . only one was verified to have a pupil defect at the time of the comprehensive examination. . ." My major concern is that clinicians around the world will generalize unwisely from a comment made by this group of highly- respected authors. Specifically, they state, "Many academic institutions. . . still teach residents and fellows to consciously look for an APD as part of any initial clinical examination. . ." The authors are clearly suggesting that is inappropriate. However, there are few tests that are as quick and as inexpensive, and whose results are so important. The presence of an afferent pupillary defect is a sure sign of significant disease. The clinical significance of its presence, then, is close to 100 percent. To test for an afferent pupillary defect properly requires that the room be dark, that the light be bright, that the patient be looking in the distance, and the light be held in each eye for three seconds. The authors have verified themselves that clearly the testing was not well done. The overwhelming majority of individuals seen by physicians have borderline findings, in which it is not clear whether the involved person is merely a variant of normal or has actual disease. To discard a test that can quickly and inexpensively solve that common dilemma seems unwise.
Sincerely,
George Spaeth, M.D.
Louis J. Esposito Research Professor
Wills Eye Institute/Jefferson Medical College
Philadelphia, PA
Conflict of Interest:
None declared