Correspondence to: Michelle S. Hovorka, Medical Anatomical Sciences, COMP-Northwest, Western University of Health Sciences, 200 Mullins Drive, Lebanon, Oregon 97355. Fax: +541-259-0201. E-mail: email@example.com
The presence of a sensory component in the first cervical (C1) spinal nerve in adult humans is not well understood. During early embryonic development, cervical nerves rostral to C1 degenerate so that no cervical nerves above C1 remain, with C1 at the confluence of embryological degeneration and persistence (Streeter, 1904; McKinniss, 1936; Pearson, 1937). While studies have documented the presence of a persistent small ganglia at the level of C1 histologically in human embryos microscopically (Streeter, 1904; McKinniss, 1936; Pearson, 1937), studies of adult human cadavers, relying on gross observations, show great variability in the presence of this ganglia persisting in adult humans at this level (Ouaknine and Nathan, 1973; Oh et al., 2002; Tubbs et al., 2007, 2009).
The embryonic studies have also consistently shown clusters of sensory neurons along the spinal accessory nerve (CN XI) in the region of C1, and often there is a connection between the C1 dorsal root ganglia and CN XI, with the accessory nerve often traveling through the C1 dorsal root ganglion (Streeter, 1904; Pearson, 1937). The continued presence of these ganglia in human adults is documented only by studies of single cases (Ingbert, 1903; Streeter, 1904). The persistence of ganglia along the C1 spinal nerve and CN XI of adult humans requires further investigation.
Recent gross examinations of the C1 region in the adult human have shown great variability in the composition of the C1 spinal nerve and its association with CN XI (Ouaknine and Nathan, 1973; Oh et al., 2002; Tubbs et al., 2007, 2009). These recent studies have found that dorsal root component of C1 is often absent, and the dorsal root ganglia cannot always be identified when a dorsal root is present (Tubbs et al., 2007, 2009). The C1 spinal nerve has also been shown to have variable anastomotic connections with CNXI, which have been classified into four general arrangements (McKenzie et al., 2009; Ouaknine and Nathan, 1973; Oh et al., 2002; Tubbs et al., 2007, 2009) (Table 1). In type I arrangements, the C1 spinal nerve is formed wholly from its ventral roots; it lacks any dorsal roots or an anastomotic connection with CN XI. Type II arrangements are “typical” in that the C1 spinal nerve is formed from dorsal and ventral roots; it has no connection with CN XI. In type III, the C1 spinal nerve is formed from both dorsal root and ventral roots and it has an anastomotic connection between the dorsal root and CN XI. In type IV, C1 dorsal rootlets are absent, but there is an anastomotic connection (Mackenzie's nerve) between CN XI and the ventral root of C1 which forms the C1 spinal nerve. Studies have found type I arrangements in 19%–44% of specimens, type II arrangements in 6%–27% of specimens, type III arrangements in 23%–38% of specimens, and type IV arrangements in 3%–44% of specimens (Ouaknine and Nathan, 1973; Oh et al., 2002; Tubbs et al., 2007, 2009).
Table 1. Percent of cases with each of the four anastomotic arrangements at the level of Cl
The present study was performed to document the location and number of afferent neuronal cell bodies at the C1 level by examining dorsal roots, dorsal root ganglia, ventral roots, and CN XI at this level, both grossly and microscopically, in adult human cadaveric specimens.
MATERIALS AND METHODS
The C1 and adjacent regions of the spinal cord was removed en bloc with attached nerves from embalmed cadavers using a posterior approach. Both males and females, ages 52–96 years, were examined. In 10 cadavers, bilateral specimens were harvested, while in eight cadavers only unilateral specimens were used due to damage or loss to the opposite side during preparation, yielding 28 specimens.
First, the spinal nerves of C1 and C2 were isolated and the brainstem and spinal cord were exposed by removing the occipital bone of the skull and the laminae of the upper cervical vertebrae. The spinal cord from the level of C3 inferiorly to the pyramidal decussations of the brainstem superiorly with all the nerve roots, CN XI, and surrounding meninges intact were then removed and were stored in 10% neutral buffered formalin for a minimum of 1 week.
Fine dissection of the C1 spinal nerve roots was preformed under a dissecting microscope with either camera lucida or digital camera attachment. CN XI and its roots at the C1 level were also removed. In cases where CN XI was attached to the C1 spinal nerve, the group was removed in such a way as to preserve the attachments. The specimens were then washed for 30 min in 50% ethanol and stored in 70% ethanol at room temperature. Following dehydration, the specimens were embedded in “paraplast.” The blocks were sectioned serially at 12 µm, and stained with hematoxylin and eosin.
The sections were viewed using an Olympus BH-2 microscope and photographed with a Jenoptik ProgRes C3 digital camera system. The digital images were captured in ProgRes CapturePro 2.5, and the ganglia were measured. Sensory neurons were identified by their morphological characteristics and the visible nucleoli were counted to determine the number of neuronal cell bodies. Volumes were calculated using the equation: V = 3/4πrxryrz, where x, y, and z represent the length, width, and depth of a ganglion, respectively.
Gross examination showed that of the 28 specimens, 11 (39.3%) had a type I anastomotic arrangement, seven (25%) had a type II anastomotic arrangement, three (10.7%) had a type III anastomotic arrangement, and seven (25%) had a type IV anastomotic arrangement (Table 1). We found no apparent relationship between the sex or age of the cadaver and the type of anastomotic relationship at the level of C1. Only six of the 18 cadavers exhibited bilateral symmetry in type of anastomotic arrangement.
Examination of the 28 specimens revealed that two contained a grossly visible dorsal root ganglion on the C1 dorsal root: one with a volume of 1.051 mm3, containing 2,850 neurons (Fig. 1A,B); the other with a volume of 1.357 mm3, containing 2,019 neurons (Table 2). Of the remaining specimens that contained C1 dorsal roots, all contained microscopically visible collections of neuronal cell bodies (Fig. 2A,B). These ranged in volume from 0.115 to 1.262 mm3 and contained between 158 and 3,096 neuronal cell bodies (Table 2).
Table 2. Ganglia within the Cl dorsal root
# Neuronal cells by visible nucleoli
Grossly identifiable ganglion
All 28 specimens contained C1 ventral roots, and histological examination revealed that none of them contained neuronal cell bodies.
The spinal accessory nerve was present in all 28 cases, and examined histologically at the C1 level in 22 of the specimens. One CN XI, with a type I anastomotic configuration, contained a grossly visible ganglion (Fig. 3) with a volume of 0.708 mm3 and containing 1,333 neurons (Table 3). The remaining 21 CN XI examined contained microscopically visible clusters of neuronal cell bodies in the region adjacent to C1, containing as many as 2,135 neurons (Table 3). In these cases, the neuronal cell bodies were located in clusters along the nerve too small to detect by gross visual inspection (Fig. 4) or hidden at the attachment points of roots and anastomotic branches (Fig. 5).
Table 3. Ganglia within CNXI
# Neuronal cells by visible nucleoli
Grossly identifiable ganglion
Neuronal cell bodies did not lie within a measurable discrete ganglion.
One of the complicating factors of the adult human C1 spinal nerve is that its composition is highly variable (Ouaknine and Nathan, 1973; Oh et al., 2002; Tubbs et al., 2007, 2009). In some cases, studies have shown grossly visible dorsal root ganglia along the dorsal root of C1, while in other cases dorsal roots are present but do not contain grossly visible ganglia (Tubbs, 2007, 2009). Still other studies have shown C1 nerves that lack dorsal roots altogether, although the ventral roots are always present (Ouaknine and Nathan, 1973; Oh et al., 2002; Tubbs et al., 2007, 2009). In some cases, there are anastomotic connections between the C1 spinal nerve roots and CN XI, and in other cases, there are none (Ouaknine and Nathan, 1973; Oh et al., 2002; Tubbs et al., 2007, 2009).
The present study, which includes a microscopic examination, also shows great variability in the composition of the adult human C1 spinal nerve. However, in our study, when C1 dorsal roots are present they always contain dorsal root ganglia, though they are not always visible by gross inspection. The ventral root, while always present, never contains a collection of sensory neurons, even at the microscopic level. We also found that, at the C1 level, the spinal accessory nerve contains collections of sensory neurons which may form a grossly visible ganglion or may only be discernable at the microscopic level as in a majority of cases. These findings are consistent with what has been shown in microscopic examination of developing human embryos (Streeter, 1904; Pearson, 1937).
No direct correlation between the size of a ganglion and the number of neurons it contains could be observed. The ganglia instead seem to lie along a continuum from those containing densely packed neuronal cell bodies to those containing diffusely spaced neuronal cell bodies. In densely packed ganglia, there are many more neurons in the same physical space with minimal neuronal processes traversing that space such as in Figs. 3 and 5. In similarly sized ganglia where the number of neuronal cell bodies is lower, such as Fig. 4, the neuronal cell bodies are spaced further apart along the nerve with a larger proportion of the ganglion volume being taken up by neuronal processes.
Many of the previous studies have shown that type III and IV anastomotic arrangements between the C1 spinal nerve roots and CN XI were the most prevalent (Ouaknine and Nathan, 1973; Oh et al., 2002). These findings are not surprising because CNXI travels through the C1 DRG during embryonic development (Streeter, 1904; Pearson, 1937). This study found type I, II, and IV anastomotic arrangements at frequencies within the range of those reported previously, while type III anastomotic arrangements were found less frequently (Ouaknine and Nathan, 1973; Oh et al., 2002; Tubbs 2007, 2009).
Our findings of clusters of sensory neurons along CN XI are also consistent with earlier findings of the presence of ganglia along CN XI during embryonic development (Streeter, 1904; Pearson, 1937). These findings may also provide an explanation for the source of sensory fibers in the human CN XI found in previous studies (Bremner-Smith et al., 1999). These sensory neurons may be involved in the proprioceptive function of some of the muscle spindle fibers of the sternocleidomastoid and trapezius muscles or may travel through anastomotic connections into cervical spinal nerves; however, additional studies would be required to determine if either of these hypotheses are correct.
In this study, only two of the ganglia identified along the C1 dorsal root and one along CN XI were visible by gross observation, the rest were visible only by histological examination. This may be due to the small quantity of sensory neurons at the level of C1. While dorsal root ganglia at the level of C2 in 13-week human fetuses contain around 20,000 neurons (McKinniss, 1936), ganglia at the level of C1 contain significantly less even in the embryonic stages (Streeter, 1904). In the present study, the ganglia identified at C1 contained 3,096 neurons or less, confirming the findings of Streeter (1904) that the dorsal root ganglia at C1 and along the spinal accessory nerve of human embryos are rudimentary and usually are not visible by gross examination but are demonstrable microscopically. Additional studies are needed to determine the distribution of these neurons, and how their presence affects the function of these nerves.
This study was done in partial fulfillment of a M.S. degree by Michelle Hovorka. The authors thank James Rhodes for his technical assistance with photomicroscopy, Karl Boehm for his work as a graduate student researcher, and Alyssa Smith for her work as an undergraduate student researcher on this project.