Neuroanatomical studies of human extraocular muscles

Detailed knowledge of the neural arrangement in the human extraocular muscles (EOM) is fundamental for the understanding of oculomotor control yet information regarding both efferent and afferent innervation is incomplete. Previous analysis of the nerve fibre distribution and estimates of the motor unit have failed to include the full range of fibre diameters and are subject to criticism. What fraction of these nerves is afferent is still unknown and the role of the sensory receptors is disputed. A variety of features observed in aged extraocular muscle spindles were described recently leading to the contention that they are probably incapable of proprioception. Other mechanoreceptors have been observed in the distal tendon, but whether or not these receptors can be equated with Golgi tendon organs is unclear. The aims of this study were, therefore, first to analyse the structure and consider the potential function of spindles in infant human EOM and to compare infant muscles with the existing knowledge of adult muscles. Secondly, tendon innervation was also studied over the same age range. Finally, total nerve and muscle fibre counts were made and motor units calculated incorporating a correction factor to allow for the estimated portion of sensory fibres. Thirty one muscle samples were obtained from 19 patients aged 1 day to 90 years. None of the patients had any history of binocular abnormalities or neuromuscular disease. The samples were fixed in 5 % glutaraldehyde, prepared using standard histological techniques and examined by light and electron microscopy. Quantitative analysis of the fibre content of 4 inferior oblique muscles displayed a reduction in the number of muscle fibres with age (correlation factor 0.969 ). The specimen obtained from the oldest subject (90 years), contained only 36% of the number of fibres found in the youngest subject (42 years). Analysis of the nerves presented a trimodal distribution of fibre diameters where the unmyelinated axons constituted a significant portion of the total. It is legitimate to include these fibres in calculation of the motor unit because of then- efferent nature, although in earlier estimates this was not done. Felder struktur fibres, known to be served by such small diameter efferents, have therefore arguably a smaller motor unit than previously assumed. In contrast to the muscles, the nerves presented insignificant variations in fibre content with age. This resulted in a variation in the size of the motor unit, from 4.75 in the youngest to 1.87 in the oldest subject. These figures were determined from nerve and muscle fibre numbers alone after subtraction of the estimated number of sensory nerve fibres. Estimates of the sensory nerves were based on the content of intrafusal muscle fibres and number of tendon receptors. Forty spindles, obtained from the distal end of 20 muscle samples, were examined in transverse serial sections . They varied in length from 47 to 683 pm (mean 323pm) and contained between 2 and 12 intrafusal muscle fibres. The majority of intrafusal muscle fibres were fragmented and/or failed to run the full length of the spindle. Most spindles had a small periaxial space and contained at least one large fibre lacking a nuclear bag or chain region. The presence of interrupted intrafusal fibres with their sensory portion disconnected from a contractile pole makes it seem unlikely that any ordered deformation of the primary sensory ending will occur during a muscle contraction. Hence, the ending may be incapable of responding to stretch. Furthermore, unmodified fibres with presumed extrafusal functional characteristics within the spindle would prevent independent action by the spindle. These factors along with other structural departures from the conventional spindle, previously found in adults and now in infants, give reason to question the muscle spindle's functional capacity. No sensory receptors were found in the distal tendon in infant material but in adults axon terminals were present in a few areas. Certain of these structures bore mechanoreceptorial features yet none of them was found to be encapsulated or presented typical Golgi tendon organ form. Their introduction in mature muscle is arguably the result of motor end plate lability and of the redundant fibres from spindles seeking new targets. The absence of tendon receptors in the infant material and the peculiar structure of the spindle suggest that muscle proprioception must be limited, at least in the early stages of life and probably is of questionable functional significance. Earlier proposals that proprioception plays a major role in development and maintenance of binocularity are inconsistent with these findings. These observations add credence to alternative explanations for eye position monitoring, in particular the efference copy hypothesis