Monday, May 20, 2019




Ajmal Khan Tibbiya College
Aligarh Muslim University
Aligarh, India


Corticosteroids are a double-edged sword. While they are useful in the management of various conditions, they also have multiple ocular and systemic side effects which may even be life threatening at times. 

In certain countries steroids are easily available over-the-counter while online pharmacies are also responsible for the unmonitored accessibility of these agents.

Ocular side effects include: increased susceptibility or reactivation of certain infections, cataract and increase in intra-ocular pressure (IOP).

Prolonged elevation of IOP can cause damage to the optic nerve leading to the development of steroid-induced glaucoma (SIG).  


SIG is a form of open angle glaucoma.

The exact mechanism responsible for elevated IOP after steroid intake is not clear.

It is most likely associated with reduced facility of aqueous outflow.

Other theories include:
Stabilization of lysosomal membranes and accumulation of polymerized glycosaminoglycans (GAGs) in the trabecular meshwork (TM). Hydration of GAGs leads to biological edema in the TM and consequently reduced aqueous outflow. Corticosteroids, which stabilize lysosomal membranes, could reduce the release of lysosomal hyaluronidase resulting in a relative inhibition of hyaluronate depolymerization.

Ultra-structurally, accumulation of basement membrane like material staining for type IV collagen is also seen in SIG. Extracellular deposition of extracellular matrix material in the TM shows 2 patterns: Fingerprint-like deposition of material in the uveal meshwork and accumulation of fine fibrillar material in the juxtacanalicular region. 

Corticosteroids cause inhibition of phagocytic properties of endothelial cells lining the TM leading to accumulation of aqueous debris.

Glucocorticoids decrease synthesis of prostaglandins, which regulate aqueous outflow. They also inhibit TM cell arachidonic acid metabolism and reduce phagocytic activity of TM cells. Dexamethasone also causes cross-linked actin network formation. However, the effect of such an alteration of cellular cytoskeleton on TM cell function is not clear. 

There is some genetic influence in the development of SIG, as several genes have been found to be associated with protective as well as damaging glucocorticoid treated TM cells. A number of genes are suspected though not yet proven to have a role in the development of SIG. These include: MYOC (previously called trabecular meshwork induced glucocorticoid response [TIGR]) gene and genes regulating alpha1 chymotrypsin, pigment epithelium derived factor, cornea-derived transcript 6, prostaglandin D2 synthetase, growth arrest specific 1, decorin, insulin like growth factor binding protein 2, ferritin light chain, fibulin-1C and others.


Steroid induced IOP elevation can occur at any age.

Children may have a severe ocular hypertensive response to topical steroids.

In a study SIG was responsible for 1/4th of all cases of acquired glaucoma in children.

According to Becker and also Armaly, individuals can be categorized into high, intermediate and low responders to topical steroids. 

HIGH RESPONDERS: Had IOP increase above 31 mmHg or 15 mmHg above baseline.
INTERMEDIATE RESPONDERS: Had IOP increase between 20-31 mmHg or between 6-15 mmHg above baseline.
LOW (NON) RESPONDERS: Had IOP less than 20 mmHg or rise of less than 6 mmHg from baseline.


Factors which increase the risk of SIG include=

1. Patient related factors: Glaucoma or glaucoma suspects have marked rise in IOP levels after several weeks of topical corticosteroid therapy. First degree relatives of POAG patients have higher risk of being steroid responders. 

2. High myopia, patients with H/O penetrating keratoplasty or refractive surgeries may have a high IOP response which gets masked due to low central corneal thickness, Ocular rigidity changes, Corneal edema and fluid accumulation beneath the LASIK flap.

3. Extremes of age: Children below 10 years of age and elderly individuals show higher steroid responsiveness.

4. Type I Diabetes mellitus and connective tissue disease (specially rheumatoid arthritis).

5. Pigment dispersion syndrome and traumatic angle recession.

6. Systemic disorders such as adrenal adenoma.


Topical: IOP elevation is more common by this route compared to systemic administration.

Periocular therapy: Long-acting repository steroids are most liable to cause rise in IOP. A patient’s previous response to topical steroids does not predict their response to periocular steroids.

Intravitreal injections: Injections of triamcinolone or depot steroids such as ozurdex can cause elevation of IOP within 2-4 weeks of the procedure.

Systemic therapy: Injections, tablets (oral), Skin preparations, Inhalational agents and other systemic steroid use is less commonly associated with rise in IOP. Use of steroids for muscle building is another danger for SIG.

The potency and strength of the steroid used are related to the IOP response which occurs. Often the response does not correlate to the dosage or duration of treatment.


Raised IOP in susceptible individuals usually occurs in a few weeks after commencement (average: 4 weeks). However, it may occur within hours or several years later.


Signs and symptoms vary with the age of the patients. Infants present with features of congenital glaucoma such as watering and photophobia. Adults show features of open angle glaucoma. SIG may be accompanied with other complications such as mydriasis and ptosis.


Primary open angle glaucoma 
Normal tension glaucoma 
Juvenile open angle glaucoma 
Uveitic glaucoma 
Glaucomatocyclitic crisis


First line of management is discontinuation of the steroid. Usually IOP returns to normal in a few days or weeks. Patients require pharmacological IOP lowering during this period.

If steroids cannot be stopped in the patient, they can be substituted with non adrenal steroids such as: rimexolone, loteprednol etabonate, fluoromethalone & medrysone.

Depot steroids need to removed by excision and clearing all steroid deposits in the area.

Intravitreal steroids may require vitrectomy to remove them.

Steroids can possibly be replaced with non-steroidal anti-inflammatory agents such as: flurbiprofen, bromfenac, ketorolac and nepafenac.

If IOP does not respond to medical treatment then Argon laser trabeculolasty can be tried. It may control IOP until the steroid-induced hypertensive effect disappears. 

Ultimately glaucoma filtering procedures or even glaucoma drainage devices are required in intractable cases.


SIG is an avoidable condition which is aggravated by factors such as easy accessibility, risk factors for IOP elevation and poor monitoring. Increasing awareness regarding this condition can lead to a reduction in the prevalence of SIG.

Thursday, May 16, 2019





Bhopal, India


Sunday, May 5, 2019

A rethink of 10-2 visual fields in early glaucoma

Conventionally, the utilization of visual field programs for analysis of the central retina, such as the 10-2 test, has been confined to patients with advanced glaucoma.  However, over the recent past, a number of studies have shown that Humphrey visual field programs for testing the retinal periphery, such as the 30-2 or 24-2, often miss central defects. A subgroup of patients has been identified which present with central visual field defects during the early course of the disease while retaining normal peripheral visual fields. The diagnosis in these patients will be missed on 24-2 or 30-2 programs unless tested on strategies focused on the central retina. This is leading to a new awareness regarding performing of 10-2 tests early during the disease process. 

To address the issue of the current role of this test in our clinical practice and whether a rethink is required regarding the application of the 10-2 test in early stages of glaucoma a review was undertaken. This review has recently been published by the US Ophthalmic Review journal.

Kindly take a look if it interests you,

With best wishes

The article is available at the following link:

Friday, May 3, 2019

Abu al-Hasan Ali Ibn Sahl Rabbani al-Tabari : A visionary

Guest author

Birjis Fatma
Ajmal Khan Tibbiya College


By 2020 India is projected to become second overall in the world with respect to the number of individuals having glaucoma. In 2010 more than 60 million people worldwide were suspected of suffering from glaucoma. This number is projected to increase to nearly 80 million by 2020. Nearly 8.4 million people were blind due to this disease in 2010 and model calculations estimate the number of those who will become bilaterally blind from glaucoma to increase to more than 11 million by 2020. 

Glaucoma is a disease known to mankind since antiquity. The term glaucoma is derived from the Greek word “glaukos” (greenish or bluish hue of the pupil). Hippocrates, in his aphorisms believed to be presented in 400 BC, used the term to describe “a kind of blindness which came with aging and was associated with a glazed look of the pupil”. However, when we critically analyze the definition of “glaukos” we can conclude that this glazed look of the pupil could be due to glaucoma but equally well possible from cataract. 

Life history of Al-Tabari:

For many years ophthalmology, like other branches of medicine, languished during the Dark Ages. Finally, in the 11th century AD, an Islamic scholar, Abu al-Hasan Ali Ibn Sahl Rabbani al-Tabari, demystified glaucoma and presented succinct clinical features which form the basis of this disease till today. This article sheds some light on this doyen of Medieval Islamic medicine. 

Just as a candle cannot burn without fire, a doctor’s zeal to treat is fueled by his spiritual core. This spiritual strength develops by devotion and concentration towards one’s field. A legendary figure who epitomizes the balance between spirituality and medicine is the great physician Al-Tabari. His contributions have left an indelible footprint in the history of medicine. His writings are a treasure which cannot be estimated for their value. His work has illuminated the world like the warm rays of the winter sun. 

Al-Tabari was born in the 8th or 9th century (usually mentioned as 838 but also 810, 808 or 783 AD) in an influential Syriac family of Merv in present day Turkmenistan. In his book Al-Radd ala al-Nasara, he wrote that he was Christian until the age of 70 and then converted to Islam. He has mentioned this in another book: Kitab-al-din wa-al-dawla. His father Sahl Ibn Bishr was a state official, highly educated and a well respected member of the Syriac community. It is mentioned that Al-Tabari’s father was a physician whose pre-eminence earned him the Syriac title of “Rabbān” which translates to “Our Master” or “Our teacher”. Sahl was the first translator of Ptolemy’s Almagest into Arabic (800). His uncle Abu Zakkār Yahyā Bin Al-Nuumaan was also a distinguished scholar and a leader of the Syriac society. 

At 10 years of age Al-Tabari accompanied his family to Tabristan (hence the suffix Al-Tabari). His early youth was spent in that region studying philosophy, medicine, religious and other aspects of natural sciences. On completing his education he subsequently moved to Iraq in 813, around the age of 30 years. In 825 he returned to Tabarīstān and became royal scribe of the Governor Māzyār Bin Qārin. He also started writing his treatise Firdaus al-Hikma (Paradise of Wisdom) at this time.

Later Māzyār was executed and Al-Tabari went back to Iraq. In Samarra he completed the Firdaus al-Hikma around 850 AD. While he was in Iraq,the Abbasid caliph Al-Mu’tasim appointed him as his divan scribe. Al-Tabari continued to work there until the death of the Caliph, upon which he returned to Samarra.

During the reign of the new Caliph al-Mutawakkil (847-861) Al-Tabari joined the court service again as a physician and courtier. Under the Caliph’s patronage Al-Tabari embraced Islam around the period of 849-850. The Caliph gave him the title of Mawla Amir Al-Muminin. Just like his birth, it is not clear in which year and at what place did Al-Tabari pass away. It is conjectured to be around 864 AD.

Al-Tabari wrote a large number of books, the exact number is unknown as some his works are no more extant. Some of his famous works include the following:
Kitab al-Ain fi al-Mualajaat (The book on Ophthalmology treatments): An enormous treatise exclusively on ophthalmology. In his own words “I have authored a distinct book completely and exclusively on ophthalmology in which I have mentioned all ocular diseases including important and unimportant, each for each temper”. Unfortunately, this book is lost.

Firdaus al-hikma (Paradise of wisdom), also known as al-Kunnāsh al-hadra, it incorporated Syriac, Greek and Indian medical systems and compendiums to form the World’s first Medical Encyclopaedia. In Book 3 (Chapter 12) of Part IV, he described ocular anatomy and diseases. 

Firdaus Al-Hikma

Al-Mu’alajat al-Buqratiya (The Hippocratic Treatments) an important book, which also deals with ophthalmology. He devoted the 4th article of the book on “Ocular diseases and their categories, benefits, creation and treatments”. In this book he mentioned “Migraine of Eye” (shaqiqat al-ayn). The condition was characterized by eye pain, a pressure sensation, opacification of ocular fluids and a dilated pupil. Apparently, this was the first reference of raised intra-ocular pressure in glaucoma. He also described 2 novel types of “ramad” (conjunctivitis). 

He also wrote: Maqala fi tib al-ain (A paper on Ophthalmology). A copy of this is presumably present in Aleppo Library in Syria.

In the Firdaus-al-Hikmat he also described the Islamic codes of ethics as personal characters of the physician which are very much contemporaneous in this era of moral degradation. This code of ethics describe the physicians obligations towards his patients, community, colleagues and assistants.

In conclusion, Al-Tabari was a visionary medieval physician who used his teachings and works to spread his extensive knowledge regarding medicine in general and ophthalmology in particular.

Sunday, April 14, 2019

Area Under the ROC curve (AUC & ROC)

Area under the curve and Receiver operating characteristic are terms often used in studies and mentioned in articles, many of them related to glaucoma. However, for many of us some of these terms are abstract and appear to give no clue as to what they mean. This post takes a look at these terms: “Area under the curve” (AUC) and “Receiver operating characteristics” (ROC). Sometimes a combination term is used:”Area under the ROC curve” (AROC).

AUC ROC is one of the most important evaluation metrics for any classification model’s performance.

An ROC curve is a graph showing the performance of a classification model at all classification thresholds. This curve plots two parameters:
  • True Positive Rate (TPR)
  • False Positive Rate (FPR)

AUC stands for "Area under the ROC Curve." That is, AUC measures the entire two-dimensional area underneath the entire ROC curve.

What is ROC?:

Receiver Operating Characteristic (ROC) is a proven yardstick to measure the accuracy of diagnostic tests. The test divides the study population into positive (diseased) or negative (non-diseased). This is done by finding a cut-off or threshold which differentiates between diseased and non-diseased (e.g. IOP= 21 mmHg). The ROC Curve tells us about how good the model can distinguish between the two conditions. A good model can accurately distinguish between the two. Conversely, a poor model will have difficulty in separating the 2 test parameters.

Let us assume we have a model which predicts whether the patient has a particular disease or not. The model predicts probabilities for each patient (in python researchers use the "predict_proba” function). Using these probabilities, we plot the distribution as shown below:

Here, the red distribution represents all the patients who do not have the disease and the green distribution represents all the patients who have the disease.

Now we pick a value where we need to set the cut-off i.e. a threshold value, above which we will predict everyone as positive (with disease) and below which will predict as negative (without disease). We will set the threshold at “0.5” as shown below:

All the positive values above the threshold will be “True Positives” and the negative values above the threshold will be “False Positives” as they are predicted incorrectly as positives.

All the negative values below the threshold will be “True Negatives” and the positive values below the threshold will be “False Negative” as they are predicted incorrectly as negatives.

Here, we have a basic idea of the model predicting correct and incorrect values with respect to the set threshold or cut-off.

To plot ROC curve, instead of Specificity we use (1 — Specificity) and the graph will look something like this:

So now, when the sensitivity increases, (1 — specificity) will also increase. This curve is known as the ROC curve.

Area Under the Curve:

The AUC is the area under the ROC curve. This score gives us a good idea of how well the model performs.

Let us take a few examples:

As we see, the first model does quite a good job of distinguishing the positive and the negative values. Therefore, in that curve the AUC score is 0.9 as the area under the ROC curve is large.

If we take a look at the last model, the predictions are completely overlapping each other and we get the AUC score of 0.5. This means that the model is performing poorly and it’s predictions are almost random.

Specificity gives us the True Negative Rate (TNR) and (1 — Specificity) gives us the False Positive Rate (FPR).

So the sensitivity can be called as the “True Positive Rate” (TPR) and (1 — Specificity) can be called the “False Positive Rate” (FPR).

So now we are just looking at the positives. As we increase the threshold, we decrease the TPR as well as the FPR and when we decrease the threshold, we are increasing the TPR and FPR.

Thus, AUC ROC indicates how well the probabilities from the positive classes are separated from the negative classes.


Unfortunately, diagnostic tests such as ROC have some limitations, such as the test may have a different sensitivity or specificity for the disease at different stages (for e.g. the test may give a different diagnostic yield in early glaucoma compared to advanced glaucoma). It may also be affected by covariates in the studied population (e.g. age, sex, rural-urban and co-morbidities). In such a mixed population, a single “pooled” ROC is often used as an average for the performance of the test. Regression analysis have also been done to assess the influence of covariates on the ROC curves.


Wednesday, March 20, 2019


The classification of glaucomas has seen a progressive change as our understanding of this condition has evolved. The anatomic, gonioscopic, biochemical, molecular and genetic basis for the classification of glaucomas has been utilized, each having its own pros and cons. With the advent of new instruments to diagnose glaucoma, classifications have also been created based on the techniques utilized. Thus, the classifications of glaucoma include those based on:

-Optic nerve appearance
-Visual field damage
-Standard HRT parameters by bagging classification trees
-Automated classification of glaucoma stages using higher order cumulant features
-Texture features using neural networks

Unfortunately, none of the classifications have been satisfactory in their attributes to describe glaucoma. This is not unexpected, since there are different mechanisms of the disease and multifactorial pathogenetic factors at work in different individuals. Presently, the classification of glaucomas based on etiology and mechanism is still applied in clinical practice, having stood the test of time over the years.

Etiologic Classification= This is based on the underlying disorder causing alteration in aqueous inflow/outflow (Aqueous humor dynamics) or Retinal Ganglion Cell (RGC)/Optic nerve damage. 
Mechanistic Classification= This is based on specific alteration in the anterior chamber (AC) angle that causes intra-ocular pressure (IOP) to rise.

These classifications have incorrectly been based on our focus on elevated IOP as the major risk factor for the development of glaucoma, excluding other factors such as vascular, genetic or biochemical mechanisms among others.


Based on the etiology, glaucomas have been divided into primary and secondary. The primary glaucomas are assumed to have the initial events leading to outflow obstruction and IOP elevation primarily in the AC angle or conventional outflow pathway. These glaucomas are not associated with known ocular or systemic disorders which could impede aqueous outflow. They are usually bilateral and probably have a genetic basis. From a therapeutic standpoint it is essential to differentiate open angle glaucoma from closed-angle glaucoma.

On the other hand, secondary glaucomas are regarded as such because of a “partial understanding of the underlying, predisposing ocular or systemic events” [Bruce Shields]. These are usually asymmetric or unilateral. While some may have a genetic basis, others are acquired. As the concepts regarding the underlying causes of the glaucomas continue to develop, the primary and secondary classifications have become increasingly artificial and inadequate.

Classification of childhood glaucomas, especially those associated with developmental anomalies of the anterior chamber angle have been dogged by overlapping and variably defined nomenclatures which frequently denotes the age of onset rather than the underlying mechanism for the glaucoma. 

Bruce Shields has recommended replacing traditional concepts with a new scheme that provides a “better working foundation for the concepts of mechanism, diagnosis and therapy that will shape the management of glaucomas for the foreseeable future”. He has classified glaucomas based on staging. According to him, glaucomas can be considered to consist of 5 stages:

Stage I: Initiating events
Stage II: Structural alterations
Stage III: Functional alterations
Stage IV: RGC and ON damage
Stage V: Visual loss

The initiating events (Stage I) are speculated to have a genetic basis. Structural changes may start occurring in the RGCs or optic nerve head (ONH), as a result of alterations in proteins in these regions. These structural alterations (Stage II) could be subtle tissue changes in the blood vessels supplying the ONH or in supportive elements of the lamina cribrosa. Or they could act through mechanisms as yet to be understood. Structural changes may lead to functional alterations (Stage III) such as reduced axonal conduction, vascular perfusion to axons in the ONH or a progressive deformity of the lamina cribrosa that may lead (alone or in conjunction with a relative IOP elevation) to glaucomatous optic neuropathy (Stage IV), which gets reflected in subsequent VF changes (Stage V).

Traditionally, glaucomas have been divided into open and closed angle.

Chronic open angle glaucoma: This is characterized by optic nerve damage in an eye which does not have evidence of angle closure on gonioscopy and there is no identifiable secondary cause. Apparently inherited susceptibilities lead to increased resistance to aqueous outflow and higher vulnerability of the ONH to the level of IOP.

Pupillary block glaucoma: Primary Angle Closure includes asymptomatic individuals with occludable angles who have not had an acute attack, as well as those who had an attack which resolved spontaneously or with treatment prior to the development of any detectable nerve damage. Primary Angle Closure Disease (PACD) has been classified by the International Society for Geographical and Epidemiological Ophthalmology (ISGEO) into:

(1) Primary Angle Closure Suspect (PACS): Such eyes have iridotrabecular contact for atleast 2700 and normal IOP, ONH and VFs.
(2) Primary Angle Closure (PAC): There is iridotrabecular contact for atleast 2700 and raised IOP and/or peripheral anterior synechiae (PAS), but with normal ONH and VFs.
(3) Primary Angle Closure Glaucoma (PACG): There is PAC with evidence of glaucomatous damage in the ONH or VFs.
(4) Acute Angle Closure Crisis: There is periocular or ocular pain, often accompanied by headache, nausea or vomiting, IOP >21 mmHg, circumcorneal congestion, corneal edema, mid-dilated pupil and shallow anterior chamber.

Developmental anomalies of AC Angles:
These represent incomplete development of structures in the conventional aqueous outflow pathway. These anomalies could be inherited or acquired during intra-uterine life and lead to elevation of IOP. In some cases the developmental anomaly is not associated with primary or systemic etiologies and regarded as primary.

Pediatric glaucomas have been classified into the following categories:

(1) Primary Congenital Glaucoma (PCG): Primary congenital glaucoma that occurs at or shortly after birth or glaucoma of any etiology that occurs in the same time frame.
(2) Primary Infantile Glaucoma: It is genetically identical to PCG but presents 1-2 months after birth.
(3) Juvenile Open Angle Glaucoma: There is no ocular enlargement; absent congenital ocular anomalies or syndromes; Open, normal appearing angles; meets the glaucoma definition.
(4) Developmental Glaucoma: This term has been used as a giant waste basket for nearly all childhood glaucomas that are not acquired immediately after birth.

Glaucomas associated with other ocular disorders
This class includes those glaucomas in which the initiating event is an abnormality of the ocular structures such as corneal endothelium, iris, ciliary body, lens, vitreous, retina and so on. Or the initiating event is a definite second ocular pathology such as tumor, hemorrhage, inflammation and so on. Secondary glaucomas are properly considered to represent those eyes in which a second form of ocular pathology has caused IOP to rise above the normal range with consequent ON damage. The second ocular pathological processes causing optic neuropathy may include=
      i.        Neovascularization
     ii.        Uveitic conditions
    iii.        Trauma
   iv.        Lens-related


Elevated IOP is the major risk factor for the development of glaucoma. However, the concept that statistically raised IOP is a defining characteristic of glaucoma has been almost universally discarded. A disadvantage of this mechanistic system is that it ignores the causes unrelated to IOP. Also, many of the glaucomas have more than one mechanism of outflow obstruction at different times in the course of disease. As a result some of the glaucomas must be classified under more than one mechanistic heading. On the plus side, the advantage of this classification is that our understanding of the mechanisms of aqueous outflow obstruction is usually more complete than our knowledge of initiating events. An understanding of the mechanism that leads to aqueous outflow obstruction is important in developing a rationale for controlling the IOP in each form of glaucoma.

Mechanisms of Open Angle Glaucoma=
The elements obstructing aqueous outflow may be located on the anterior chamber side of the trabecular meshwork [TM] (pretrabecular mechanisms); in the TM (trabecular mechanisms) or distal to the meshwork, in the Schlemm’s canal or further along the aqueous drainage system (post trabecular mechanisms).

Angle closure glaucoma mechanisms=
Angle closure mechanisms are the ones which cause apposition of the peripheral iris to the TM or peripheral cornea. The peripheral iris may be pulled (anterior mechanisms) or pushed (posterior mechanisms) into this position. In anterior mechanisms usually a contracting membrane in front of the iris pulls the iris towards the TM/peripheral cornea. It can also be caused by consolidation of inflammatory products in this area.
In posterior mechanisms pressure behind the iris, lens or vitreous causes the peripheral iris to be pushed into the anterior chamber angle. These mechanisms can occur with or without pupillary block. Pupillary block variants include pupillary block glaucoma in which there is apposition of the mid-periphery of the iris and the lens, thus blocking the egress of aqueous anteriorly through the pupil. The peripheral iris balloons in the form of “iris bombe”. The functional apposition in these patients is due to a genetically influenced configuration of the anterior segment of the eye. Such appositions may also be seen in lens-induced mechanisms such as phacomorphic glaucoma or ectopia lentis. Pupillary block can also occur from posterior synechiae. The “pushing” mechanisms can also occur without pupillary block such as ciliary block (malignant glaucoma), lens induced, forward shift of vitreous following lens removal, intraocular tumors, cysts of uveal tract, retrolenticular tissue contraction as in retinopathy of prematurity or persistent fetal vasculature.

Developmental anomalies of the AC Angles=
These represent incomplete development of structures in the conventional aqueous outflow pathway. Examples of these include: congenital glaucoma, Axenfeld-Reiger syndrome, Peter anomaly and iridocorneal adhesions.