Effect of Osteopathic Cranial Manipulative Medicine on Visual Function

To the Editor:

We read with interest the research article by Sandhouse et al titled “Effect of Osteopathic Cranial Manipulative Medicine on Visual Function.”¹ The authors conducted this study to determine the effect of osteopathic cranial manipulative medicine (OCMM) on multiple functional visual parameters both before and after management of sphenobasilar synchondrosis (SBS) dysfunction. Few studies have been conducted in this area, especially with a blinded randomized prospective design, which is a strength of this study. However, we have concerns regarding the methods.

The authors list several testing procedures. The first measurement was “best-corrected distance visual acuity.” Participants were asked to read the Early Treatment of Diabetic Retinopathy Study chart, the criterion standard for checking visual acuity (VA) in research studies, either uncorrected or using their current corrective lenses. However, best-corrected VA is normally measured after a manifest refraction using a phoropter to ensure that the participants’ actual best potential vision is recorded. Some participants may have had latent hyperopia or a small degree of myopia that, with correction, could have been refracted to achieve better VA. Likewise, they would have performed differently in all of the measurements tested in the study. This point is extremely important, as not only do many people have small, or even large, degrees of uncorrected refractive error, but also the VAs of these participants were measured weeks apart, so any slight change in ocular surface disease could have changed the refractive error. Also, the only way to definitively determine latent hyperopia, especially in a young population, is through a cycloplegic refraction, which was not performed in this study. Therefore, the measurements obtained in the study cannot be considered “best-corrected.” (Of note, there was a slight, but not statistically significant, increase in VA more so in the control group than the treatment group.) Furthermore, without a manifest refraction, participants may not have actually met the inclusion criteria, as some may have had additional refractive errors that would have excluded them from the study.

In addition, the authors¹ claim that all of the included patients had no irregular astigmatism, as deduced through a patient questionnaire and noncycloplegic retinoscopy. However, taking the participants’ word regarding the topography of their cornea is flawed. Also, in current practice, irregular astigmatism is diagnosed based on topography of the cornea using advanced diagnostic equipment. Although performing cycloplegic retinoscopy may identify larger degrees of irregular astigmatism, it will not pick up subtle changes. Such slight changes in corneal curvature could have altered many of the measurements that the authors performed.

A proper eye examination consists of a dilated slit lamp examination of both the anterior and posterior segments of both eyes. Using the slit lamp, the ocular and periocular tissues are examined with 5× to 24× magnification, which is standard in diagnosing almost all ocular diseases. However, this part of the examination was not performed in the study.¹ As most eye care professionals will attest, many patients have undiagnosed or undertreated ocular surface disease, which has a major effect on refractive error and VA. If the participants in this study had undiagnosed ocular disease, it would have affected many of the visual measurements taken by the authors.

The other visual measurements that achieved statistical significance were the near point of convergence (NPC), Donder pushup, and stereoacuity. Most eye care professionals do not test for stereoacuity that is better than 40 seconds of arc, which is what almost all participants had in this study, nor is there a differentiation made clinically between intervals of 10 seconds of arc. The change in stereoacuity noted in the study is of no clinical significance in the day-to-day practice of eye care.

Regarding vergence, Lee and Galetta² commented on a study that assessed football players’ change in visual function after head trauma. They discussed that the full testing of vergence—including peak velocity, fusional amplitude, and divergence—was not performed, which is similar to the study by Sandhouse et al.¹ They also stated that the difference in NPC was small, which is also similar to the study by Sandhouse et al,¹ as the average NPC was within normal limits throughout each group. Lee and Galetta² also discussed the confounder of fatigue relative to poor sleep, excessive near work, and drug and alcohol intake that were not assessed before each visit, all of which can affect vergence measurements. These confounding factors, which were also not specifically assessed by Sandhouse et al,¹ limit the reliability of the measurements taken by the authors. Additionally, without a cycloplegic refraction, it is impossible to perform any of these tests because they must all be performed with the patients’ full refractive correction.

The final visual measurement discussed was pupil size. The most commonly used method of measuring pupils for research purposes is by using a pupilometer, which is a proven, standardized method that will give the most objective result. Sandhouse et al¹ measured pupil diameter by putting a ruler close to the participants’ eyes. The pupillary sphincter and dilator muscles are extremely dynamic. Any stimulation, whether a sympathetic reaction due to fear of a ruler coming close to the eye, accommodation (either spasm or momentary), a slight change in the luminance of the bulb, or a flicker in the bulb lasting under a second while the measurement is being taken, can change the size of the pupil. Thus, the change in pupil size of less than 0.3 mm reported by Sandhouse et al¹ is of no clinical significance, especially in young, healthy participants, and we believe it is likely not statistically significant.

In addition, it is unclear how long after the intervention that the reassessment of vision occurred. With the application of osteopathic manipulative treatment, it is important to measure the effects multiple times after manipulation. In the study by Sandhouse et al,¹ multiple measurements would be needed to elucidate whether there was any immediate change in the visual system resulting from OCMM, how long it lasted, and at what rate the changes regressed without further treatment.

We also question the choice of OCMM as the intervention. The visual system involves numerous neurons and pathways that span the majority of the central nervous system and parts of the peripheral nervous system. Thus, OCMM can have tremendous effects on different parts of the visual system. The authors stated, both in this study¹ as well as in their pilot study,³ that they attempted to show that treatment of the SBS will affect the visual system, but they did not state why they chose OCMM. Having an understanding of the effect on the inner workings of the cranium and central nervous system and how OCMM specifically will affect visual function is necessary for the understanding of the successes and failures of studies, as well as assisting in the construction of new studies.

With the numerous limitations that the authors discussed, including some of those we have mentioned, we question the validity of this study's results.¹ Well-constructed studies are needed to determine OCMM's full potential on the visual system. There is great potential for future studies of OCMM on the visual pathway as it is connected to the majority of the central nervous system and portions of the peripheral nervous system. In this era of evidence-based medicine, all studies must be conducted by adhering to the highest standards of rigor and understanding.