Secondary Logo

Share this article on:

Diagnostic Dilemmas in Fourth Nerve Palsies

Siatkowski, R. Michael MD

doi: 10.1097/WNO.0000000000000560
Invited Commentary

Dean McGee Eye Institute/Department of Ophthalmology, University of Oklahoma, Oklahoma City, Oklahoma.

Address correspondence to R. Michael Siatkowski, MD, Dean McGee Eye Institute/Department of Ophthalmology, University of Oklahoma, 608 Stanton L. Young Boulevard, Oklahoma City, OK 73104; E-mail:

The author reports no conflicts of interest.

Supported in part by an unrestricted grant from Research to Prevent Blindness, Inc. New York, New York.

The article by Drs. Ivanir and Trobe (1) is an excellent example of the type of clinical research which has resulted in so many important contributions to neuro-ophthalmology, i.e., careful observation of clinical signs over an extended period, followed by hypothesis development and assessment using one's clinical database. Although the potential pitfalls of missing or nonstandardized data, and lack of ability to control for various confounders can never be eliminated, such retrospective analyses still often provide many useful clinical pearls.

What are the main conclusions of this report, and are they consistent with what we already know about fourth nerve palsies?

  1. A majority (58%) of cases of congenital palsies demonstrate hypertropia (HT) greater in upgaze than downgaze, but 29% were worse in downgaze. This certainly makes sense and meshes with the clinical experience of the strabismologist community. The most likely explanation for this variability is that congenital pareses come in many different varieties and levels of severity based on factors such as the laxity of the muscle tendon, the point of insertion on the globe, development of amblyopia and fixation preference, and long-term muscle adaptations of the ipsilateral inferior oblique and superior rectus or the contralateral inferior rectus muscles, depending on which eye is fixing. Any or all of these can combine to make the HT larger in various positions of gaze. History and old photograph review are still crucial supporting pieces of information for making the diagnosis of congenital fourth nerve palsy.
  2. Measurement of cyclotorsion and vertical fusional amplitudes were not reliable distinguishers between congenital and acquired palsies. One obvious reason for this is the lack of data in this retrospective series but, again, the factors mentioned above play a role as well. In addition, incomplete paresis affecting the anterior vs posterior portions of the superior oblique muscle will affect torsional and vertical position of the eye, respectively. Our limited ability to quantitate the degree of severity of a partial paresis will always make this analysis difficult.
  3. The HT was worse in downgaze than upgaze in 100% of microvascular ischemic palsies. This is a really important point for clinicians to understand. It is rarely, if ever, explicitly stated (and previously not proven) but is true. These are typically patients who have normal muscle anatomy and suffer an acute mononeuropathy. By definition, the deviation must be largest in the field of action of the affected muscle, i.e., downgaze. Similarly, the monoparesis almost always begins to improve before secondary contracture or adaptation of other muscles can occur, leaving the presentation pattern intact until spontaneous resolution occurs.
  4. The HT was worse in downgaze in 100% of palsies due to tumor, and three-quarters of those due to trauma. This conclusion is likely true acutely, but less generalizable in longer-standing deviations because it does not take into account the issue of length of time from onset to presentation/diagnosis. Although like microvascular cases, these patients presumably have normal anatomy, their conditions may not spontaneously improve even partially, allowing for various muscle contractures or adaptation. As an example, I recently saw a patient who had neurosurgery for an intracranial vascular lesion. Before the procedure, he had no diplopia, but afterward developed a left fourth nerve palsy, which did not improve over 6 months. At that time, he had –2 superior oblique underaction, no inferior oblique overaction, and a left HT of 6 prism diopters (PD) in primary gaze and 10 PD in downgaze (behaving as predicted by the report of Ivanir and Trobe). He declined surgical intervention at that time and, 6 months later, began to develop over-elevation in adduction of his left eye. His primary deviation increased to 8 PD with a similar deviation in upgaze, presumably due to inferior oblique muscle length adaptation. One year later, he had developed a deficit to depression in abduction, presumably due to focal contracture of the ipsilateral superior rectus. At that time, his HT was 14 PD in upgaze, 10 PD in downgaze, and 8 PD in primary. This patient initially behaved as described in the manuscript but over 18 months changed patterns. The authors arbitrarily set a time cutoff of 24 months in their series, but this case was still within the period in which this patient “disobeyed” their rules. In addition, if trauma (surgical or otherwise) or tumor also produces loss of vision and force fixation with the paretic eye, similar changes in pattern of deviation may occur as well, perhaps in a much more rapid time course than the patient just described.

There is one other factor that limits generalizability of these data, namely methods of measurement of the angle of squint. To be consistent, all patients should have full deviations (phoria + tropia) measured through alternate prism and cover testing, and perhaps even after a period of monocular occlusion in those with strong fusional amplitudes or small deviations. In addition, upgaze and downgaze measurements should be compared only against those in the exact same position of gaze, e.g., up and right/left cannot be compared against straight up. The Methods section of the article does not indicate these were the case and, thus, lends concern about comparing apples to oranges.

Despite these issues, I think this is a well-conceived, well-written, clinically useful study. The first 3 conclusions listed above are likely correct and generalizable. The fourth could be refined in a prospective trial using standard measurement techniques, adding cyclotorsion data, and comparing patterns in untreated patients over longer periods of time. Until those data are available, we should exercise diagnostic caution with respect to the pattern of squint in subacute and chronic fourth nerve palsies.

Back to Top | Article Outline


1. Ivanir Y, Trobe JD. Comparing hypertropia in upgaze and downgaze distinguishes congenital from acquired fourth nerve palsies. J Neuroophthalmol. 2017;xx:xx.
© 2017 by North American Neuro-Ophthalmology Society