Vergence after TBI (traumatic brain injury)

Timothy C. Hain, MD•Page last modified: May 12, 2018


Vergence eye movements are used to track objects that move in depth in one's binocular visual field to attain and maintain a fused and single percept.

Traumatic brain injury (TBI) is caused by an external insult to the head following motor vehicle accidents, falls, assaults, etc.


There is a small literature that suggests that disturbance of ocular vergence is common after traumatic brain injury and furthemore is responsible for a wide range of vision disturbance, as well as some suggest that visual disturbances can also lead to impact on balance and cognitive function. The purpose of this page is to provide a discussion as to the validity of these ideas and the evidence that is available with respect to these assertions.

Prevalence of oculomotor problems post TBI:

Oculomotor in TBI

The figure above is from Thiagarajan et al (2011), and reports the frequency of "oculomotor dysfunction" in a "Brain Injury" clinic population of Ciuffreda et al (2007). Rather amazingly, this figure suggests that almost 60% of this population has a disturbance of vergence. These patients largely had convergence insufficiency (42.5%). According to the methods section of this paper, these patients were drawn from patients referred from rehabilitation hospitals, with the diagnosis of "TBI". The paper does not say whether these were minor TBI -- e.g.. ping-pong ball falling on top of head, or severe -- e.g. falling on one's head from 2 stories. Lacking basic information about who was being studied here, makes this figure difficult to interpret.

Thiagarajan et al (2011) next provide a table again enumerating the frequency of oculomotor disturbances in other studies were all of severe trauma victims, from the military.

Because these studies again were of very severe head trauma, that occured in military settings, it is difficult to ascertain the relevance and in particular how these findings "scale" with the intensity of head injury. We find it odd that the paper of Thiagarajan et al, which concerned "mild TBI", largely discussed data from severe TBI in their review.

Alvaraez et al (2012) reported "Convergence insufficiency without simultaneous visual or vestibular dysfunctions was observed in about 9% of the visually symptomatic TBI civilian population studied." This is a far more reasonable figure -- as it both is far lower than that of Thiagarajan, and also it is confined to individuals with "visual symptoms". Alvarez et al did consider the severity of TBI, and noted that they were a mixture of inpatients and outpatients, and that the inpatients were likely more severely affected with pathology in the moderate to severe TBI range. Alvarez et al noted that they found less convergence problems in their population, some of which were rather severe, than the previously mentioned results in military populations, and suggested that one explanation might have been differences in the degree of convergence problem accepted as being abnormal. In other words, they were stricter. They also noted that the general population, without a history of head injury, has somewhere between 3.5 to 8% CI.

Alvarez et al noted that the CITT (convergence insufficiency trial) required CI as an NPC of 6 cm or more. Or in other words, someone sees double at distances greater than 6cm. Of course, one should not see double at distance, so this means that the near point has receded. Alvarez also used "Sheard's criterion", which is somewhat more stringent. According to Lavich et al, many authors use 10 cm to define CI.

Measurements of Vergence

Thiagarajan et al (2011) reports an immense number of measures of ocular function. Table 2 contains 15 separate measurements, taken from "unpublished data" of themselves and other authors. Table 3 contains another 9 measures of "dynamic" parameters. The method of defining "mild TBI" is not reported in their method section, so we essentially do not know how severe the TBI was in these patients. This makes this data meaningless.

In the same paper, they state that 5 static parameters "revealed a significant difference" between the mTBI (of whom we know nothing), and the normal groups. They stated that these differences were "significant", but the p value was set to 0.05. For a "fishing expedition" like this paper, because random things happen, it is best to divide the "p" value by the number of measures -- in this case, 24. Using the Bonferroni correction, this would almost certainly make none of their parameters significant.


It is very reasonable that severe head injury should disrupt visual function, including vergence. This appears well established.The literature reviewed herein concerns severe head trauma, or undefined doses of head trauma. It does not establish the scaling between head trauma and vergence disturbance.

It seems quite obvious that smaller amounts of head injury will result in less often and less severe disturbances to visual function. A ping pong ball falling on one's head is not the same as a fall from 2 stories. Much more work is needed.

It also appears that the criteria used to define CI vary. According to Lavrich, 10cm is used by many authors to define CI. We think it would be most helpful to have age norms and percentile ranges.