Skip to main content
NCBI home page
Journal of Anatomy logoLink to Journal of Anatomy
. 2018 Nov 8;234(1):33–42. doi: 10.1111/joa.12898

Pre‐ and postnatal development of the otic ganglion in humans

Dave Bandke 1,†,✉, Konstantin Ebauer 2,†, Alexander Ebauer 3, Serge Weis 1,2
PMCID: PMC6284433  PMID: 30411352

Abstract

Only a few papers exist dealing with the development and aging of the autonomic nervous system – and even rarer are studies that investigated the otic ganglion. Using a special trepan, we removed and investigated 172 samples from 86 corpses, ranging from 20 weeks of gestational age (GA) to 95 years of age. The aim of the study was to measure different morphometric parameters of the ganglionic neurons in order to study age‐related changes from early development until old age. Fetuses show the highest numerical density of neurons. Then, in the first years of life, a rapid growth of the cytoplasm takes place, which is the main reason for the neuronal growth and the increase of the general size of the otic ganglion at this age. Also, the number of satellite cells increases till puberty. In adults, the parameters are relatively stable over decades and decrease slowly, in contrast to the steep increase in the first years of life. Moreover, neuronal degeneration, storage of pigments, neuro‐axonal dystrophy, and lymphocytic infiltrates increase with age.

Keywords: aging, autonomous nervous system, morphology, otic ganglion, parasympathetic nervous system

Introduction

There is little data available about the pre‐ and postnatal development of the cranial parasympathetic ganglia in general – and these are mainly based on animal studies.

Anatomy

The otic ganglion is located in the infratemporal fossa inferior and medial to the oval foramen; seldom is it found in a bony recess at the border of the oval foramen. The ganglion lies inferior to the ala majoris ossis sphenoidalis and superior to the levator veli palatini muscle. Laterally, the ganglion reaches the medial surface of the mandibular nerve, for which it forms an ‘adhesion‐ and distribution’ center. Medially to it lies the cartilage of the auditory tube. Anterior to it runs the medial pterygoid nerve (as a continuation of the motor root of the trigeminal ganglion). The medial meningeal artery runs posterior to the otic ganglion.

Four fiber qualities are traditionally described to be present in the otic ganglion: parasympathetic, sympathetic, motor, and sensory fiber bundles. However, only parasympathetic fibers are interconnected and, therefore, define the function of the otic ganglion. It receives its parasympathetic fibers from the inferior salivatory nucleus. For further representation of the topography of the otic ganglion, including a photo‐documentation, we refer to newer studies (Siessere et al. 2008; Lovasova et al. 2013; Senger et al. 2014).

Short historical sketch of the morphology of the otic ganglion

The otic ganglion was first described by Arnold in 1829 as a part of the sympathetic nervous system (Arnold, 1829), nowadays described as the autonomic nervous system; the subdivision into ‘sympathetic’ and ‘parasympathetic’ nervous system, as we know it, did not exist until 1921.

Retzius was then the first who visualized neurons of the otic ganglion using fiber teasing preparations. He could only show convincing pictures of multipolar neurons from cats and rabbits, not from humans (Retzius, 1880). Using metallic silver impregnation techniques, Müller & Dahl (1910) presented neurons and their ramifications in animals, but they could only show crown cells in humans, i.e. neurons with small short skein‐like bended intracapsular dendrites. A few years later, Riquier was able to demonstrate neurons in human otic ganglion tissue using silver impregnation techniques of Cajal and method II of Donaggio (Riquier, 1913). These neurons were multipolar, indicating that the otic ganglion belongs to the sympathetic nervous system.

Tanaca described the otic ganglion with its connections in 30 alcohol‐fixed skulls and gave a detailed description of its form and size (Tanaca, 1932). However, that study was forgotten since the connections of the otic ganglion were published in later papers as new descriptions.

The development of the otic ganglion was then accurately described by Anders & Kautzky (1955). Similar to other ganglia, the otic ganglion develops from neuroblasts belonging to the cranial ganglionic line. It can be observed as early as in 32‐ to 34‐day‐old embryos; 40‐ to 43‐day‐old embryos had ganglia which were filled with immature cells (Anders & Kautzky, 1955).

Based on electron microscopic studies, Sangiacomo subdivided the neurons into two types, dark and light, based on the organization of the endoplasmatic reticulum (Sangiacomo, 1969).

Since the 1980s, there have been numerous studies, mostly on rats, that investigated the nerval connections of the otic ganglion. Although they are rarely mentioned in textbooks, they may become clinically relevant.

Postganglionic fibers can be divided into four large topographic areas: (i) parotid gland, (ii) vessels of the orofacial and nasal region, (iii) external carotid artery with its branches, and (iv) internal carotid artery with branches.

  • The connections to the parotid gland are well known and can be looked up in textbooks; thus, short explanations will follow on the remaining anatomical areas.

  • In animals, most of the branches of the mandibular nerve carry postganglionic fibers of the otic ganglion. In that way, the otic ganglion also participates in the innervation of the pulpa of the mandibular teeth (Segade & Suarez Quintanilla, 1988), and vessels of the lower lip and the facial skin (Kaji et al. 1988; Izumi & Karita, 1992; Kuchiiwa et al. 1992). Also, the nasal rat mucosa receives partial parasympathetic efferences from the otic ganglion (Grunditz et al. 1994).

  • The middle meningeal artery and the superficial temporal artery in rats are innervated by the otic ganglion (Uddman & Edvinsson, 1989; Uddman et al. 1989). The superficial temporal artery most likely receives efferences from the otic ganglion through the auriculotemporal nerve. This is clinically interesting because the superficial temporal artery is involved in migraine attacks.

  • The postganglionic fibers of the otic ganglion innervate the blood vessels of the circle of Willis, with the exception of the vertebrobasilar artery (Suzuki et al. 1988, 1990; Edvinsson et al. 1989; Shimizu, 1994). Ruskell found 15–24 outgoing rami dorsales that carried marrow and marrowless fibers. They run through the oval foramen back to the inside of the skull and form the cavernous sinus plexus with sympathetic fibers from the internal carotid plexus. The cavernous sinus plexus is the place where all parasympathetic fibers from otic and sphenopalatine ganglia as well as smaller ganglia switch on the internal carotid artery (Ruskell, 1993). This anatomical position is important for understanding the pathology of syndromes with retroorbital pain (cluster headache, ophthalmoplegic migraine).

Also based on these findings, the sphenopalatine ganglion has been identified very recently as a therapeutic target in severe cluster headache (Narouze, 2014; Schoenen, 2015; Akbas et al. 2016; Barloese et al. 2016; Lainez & Marti, 2016).

Furthermore, the co‐existence of different neurotransmitters in the neurons of the ganglion has also been known for three decades (Sharkey & Templeton, 1984; Leblanc et al. 1987; Gibbins, 1990; Grunditz et al. 1994). Recently, pituitary adenylate cyclase‐activating peptide (PACAP) has been observed in the otic ganglion and identified as a neuropeptide that may play a major role in primary headaches (Edvinsson et al. 2018).

To our knowledge, no study has systematically investigated age‐related changes in the human otic ganglion in a larger cohort. Therefore we examined 172 specimens of 86 human beings in order to discover histological and morphologic changes. Our aim was to answer the following question more adequately: Which macroscopic and light microscopic changes are to be expected at which age?

Material and methods

A total of 172 specimens were taken from 86 individuals ranging in age from a 20‐week‐old embryo to a 95‐year‐old woman (Table 1). Samples were taken 12–24 h postmortem using a specially developed trepan. Our first experiments with classical trepans with a serrated edge showed that these instruments work like a core bore and cause massive tissue damage due to axial shear forces. With our trepan, we were able to remove the otic ganglion in a fast and safe way without damaging tissues or disfiguring the corpse (Figs 1 and 2). After craniotomy and removal of the brain, the root of the trigeminal nerve and the trigeminal ganglion with its three trees were identified. By gently lifting up the trigeminal ganglion, the main branch of the trigeminal nerve, i.e. the mandibular nerve, was identified and separated from the ganglion. While the head was fixed, the skull base was trepanated at the level of the oval foramen in such a way that the instrument reached the pharyngeal vault. Then, the trepan was removed under gentle rotation. Because of the grinding angle of our trepan (facing outwards 20°), the tissue, captured inside the instrument, did not undergo compression and the bone splintered only minimally. This fine macroscopic preparation was performed using a binocular magnifying lens. In all investigated specimens, the otic ganglion was found at the typical topographical location. The fresh otic ganglion was reddish gray and became more yellowish at higher ages.

Table 1.

Overview of average perikaryon size and nucleus size plus standard deviation in 172 specimens retrieved from 86 cadavers

Perikaryon size (μm2) Nuclear size (μm2) n (=172)
Fetuses (GA 20–40) 14.71 ± 1.21 6.78 ± 1.90 12
Newborns (0–4 weeks) 18.73 ± 2.30 7.50 ± 2.22 14
Babies (1–12 m) 28.20 ± 3.39 9.81 ± 2.19 22
Infants (1–5 years) 29.45 ± 3.52 10.74 ± 2.22 12
Schoolchildren (6–17 years) 34.26 ± 2.54 11.84 ± 1.21 20
Young adults (18–14 years) 38.40 ± 0.41 13.32 ± 0.21 12
Adults I (25–44 years) 38.64 ± 3.46 13.44 ± 1.17 26
Adults II (45–59 years) 42.80 ± 3.50 15.34 ± 1.21 14
Elderly people I (60–74 years) 37.91 ± 0.63 13.62 ± 0.19 20
Elderly people II (≥ 75 years) 36.25 ± 0.50 12.28 ± 0.19 20
Open in a new tab

Figure 1.

Figure 1

Open in a new tab

The shown trepan was used to punch the otic ganglion. Because of the punching, there is no rotation and thus rotation‐related damage is avoided. After removal, the tissue is pushed out through the side slit (right side of the image). It undergoes almost no compressive forces due to the outward‐pointing grinding, which makes this trepan special.

Figure 2.

Figure 2

Open in a new tab

The extraction site is shown with a ‘tube’ on a prepared skull. On the contralateral side, the target area is marked in red. The tissue extraction proceeds in a caudalward direction from the skull base and includes part of the skull bone with parts of the mandibular nerve.

The preparations were divided into two groups: The left otic ganglion was used for silver impregnation stainings, and the right one was used for further histological stainings. Depending on the type of staining procedure, samples were fixed either with 10% formalin or with Karnua's solution. Cryostat sections of 30–40 μm thickness were used for silver impregnation staining (Bielschowsky's with Rasskasowa method). Other stains –hematoxylin‐eosin (HE), Elastica van Gieson (EVG) and Periodic‐acid Schiff (PAS) – were performed on paraffin‐embedded tissues which had been cut into sections of 5–7 μm thickness.

The numerical density of neurons within a ganglion was determined following the rule of the unbiased test grid (Gundersen, 1977) and was calculated as the number of cells per square millimeter (n mm−2). The size of neurons was determined using the point counting method (Weis, 1991). The following parameters per individual cell were evaluated:

  • profile area of the nucleus

  • profile area of the cytoplasm

  • profile area of the cell (i.e. the sum of nucleus and cytoplasm)

  • profile area of the space delineated by the satellite cells (satellite ring)

  • profile area of the empty space between the satellite ring and the cytoplasm, i.e. ring minus cell (satellite–cytoplasm interspace),

  • number of satellite cells

Furthermore, with regard to the profile areas, the following ratios were calculated as percentages:

  • nucleus/cytoplasm (RA1)

  • nucleus/cell (RA2)

  • nucleus/satellite ring (RA3)

  • cytoplasm/cell (RA4)

Since the number of satellite cells per cell was also counted, the following ratios were calculated and expressed as percentages:

  • satellite cells/neuronal nucleus (SAT1)

  • satellite cells/neuronal cytoplasm (SAT2)

Level of significance between age groups was tested using Student's t‐test and the non‐parametric Mann‐Whitney's U‐test. Correlation analyses were also performed using the Statictical Package for the Social Sciences (SPSS) (version 18).

Results

Qualitative analysis

Fetuses (gestation age 20–40 weeks)

The fetuses had fully developed ganglia. Microscopical examination revealed a loosely structured, well‐vascularized capsule of connective tissue bounding the ganglion. A few flimsy and poorly myelinated nerve fibers formed a fine network. The entering and emerging nerve fiber tracts contained Schwann cells in high density. A fine vascular network (arterioles, venules, and capillaries) was mainly present in the capsule and partly in the parenchyma. Nerve cells were in different stages of development and were densely arranged. Depending on their grade of maturation, they were classified as syncytium‐like groups of neurons, neuroblasts, and immature neurons. Six‐month‐old fetuses most often exhibited neurons with joint capsular plasmodium. Up to the 8th month of gestation, neurons showed a tendency to disentangle; in such cases, mainly larger neurons had their own, satellite‐studded capsules.

Newborns (0–4 weeks)

The capsule was well defined and send out fine trabeculae into the ganglion center. The morphological appearance of nerve fiber bundles and Schwann cells was similar to that observed in fetal tissues. Syncytium‐like conglomerates of cells with joint protoplasm had a maximum number of four nuclei and were only discovered in two cases. Apolar neuroblasts were arranged in tracts and nests, respectively, and were mainly found in peripheral parts of the ganglia. The most common elements of nerve cell clusters were immature neurons. The size of immature neurons increased in the course of development (from fetus to newborn). There were often four or more cellular processes originating from immature neurons. In week 4 after birth, one‐half of neurons had their own capsules formed by independent satellites. One flimsy neurofibrillary tangle was visible.

Babies (1–12 months)

All elements of the ganglia had matured: capsules and septa were thickened, and the number and size of blood vessels were increasing. Nerve cells and glial cells were growing in size. Nerve fibers were widening: the ratio of myelinated to non‐myelinated fibers was increasing. No syncytium‐like cell conglomerates were visible. Only few nests of neuroblast‐like cells were present. These nests were mainly at the fringe; this pattern of distribution remained with individuals of higher age. From 12 month of age onward, nearly each nerve cell had its own smooth capsule. Sometimes processes entangled with others were forming some sort of ‘fenestration’. The first cases of protoplasmic lobules occurred. There was also one case with a neuron showing signs of degeneration (argyrophilia and irregular‐shaped perikaryon in a 5‐month‐old infant). We also observed two neuro‐axonal dystrophy (NAD)‐lesions in one case (see the end of discussion).

Infants (1–5 years)

The ganglionic capsules thickened. The number and caliber of blood vessels were slightly increasing. Nerve fibers were thickened and more myelinated fibers were prevailing. The number of satellite cells increased also: bigger neurons were surrounded by up to 11 satellite cells. At the age of 5, most nerve cells had differentiated into multipolar neurons with numerous processes.

Schoolchildren (6–17 years)

Schoolchildren did not show any features distinct from elder infants. From the onset of puberty on, connective tissue was rich in collagen fibers. There was a distinct neuronal capsule around big neurons. Bigger neurons were surrounded by up to 15 satellite cells. Stalks of protoplasm were increasingly visible in this age group. There was a slight increase of degenerated neurons (argyrophilia, in some cases formation of vacuoles) which was strongly dependent on each individual case. Some of the microscopical slides contained even three to four equally distributed NAD lesions (neuro‐axonal dystrophies).

Young adults (18–24 years)

The structure of ganglia changed only slightly. Structurally, the entering and leaving nerve bundles remained the same. The number of satellite cells around big neurons increased only slightly. The composition of neuron cell types showed no changes compared with the previous age group. Very rarely, apolar neuroblasts were found. Development of multipolar neurons became more complex. The amount of neurons with stalked protoplasmic lobes increased.

Adults I (25–44 years)

There were no apolar neuroblasts. In the third decade of life, stalked protoplasm increased tremendously, but varied from one individual to the other. Nerve fibers thickened slightly, but there was a higher incidence of varicose swellings of unmyelinated nerve fibers. We also found lymphocytic infiltrates in two cases (15%) for the first time.

Adults II (45–59 years)

The ganglion size remained the same. In this age group, we found the neurons with the highest number of satellite cells (over 30 satellite cells). Microscopically, swellings of unmyelinated nerve fibers were apparent. Some arteries showed a hypertrophic intima media. There were no lymphocytic infiltrates.

Elderly I (60–74 years)

There was hardly any change to the previous age group. Many vessels, especially arteries, showed intimal fibrosis with lumen stenosis. Lymphocytic infiltrates were found in two cases (20%). Some nerve fibers had varicose, spindle‐shaped thickenings and argyrophilic widening; this also applies to entering and leaving nerve fibers.

Elderly II (>75 years)

Most ganglia appeared yellowish. The capsule, septa, and intercellular connective tissues showed a significant increase in collagen fibers. Many arteries showed thickening of their wall; occasionally with lumen obliteration. There were two cases with lymphocytic infiltrates (20%). Nerve fibers presented with argyrophilic widening, thickening and occasionally fragmentation. Atrophy of neurons was evident (see also Table 1, further reduction in the size of perikaryon and nucleus).

Quantitative analysis

Table 1 gives an overview of the perikaryon and nuclear size plus standard deviation for each age group.

Numerical density of neurons

The peak of the numerical density occurred in fetuses. Neuronal density decreased rapidly until the onset of puberty because of perikaryon growth (r = −0.63; P < 0.00001; see Fig 3a). After the age of 30 years, neuronal density decreased further, due to neuronal loss and consecutive proliferation of connective tissue and glial cells. Numerical density was significantly negatively correlated with age (r = −0.70; P < 0.001; see Fig 3b).

Figure 3.

Figure 3

Open in a new tab

Different graphs showing the data for the numerical density of nerve cells (a,b), the mean neuronal cell size (c), the mean neuronal nucleus size (d), the mean neuronal cytoplasm size (e), the mean number of satellite cells (f), the satellite cells vs. neuronal nucleus size relationship (g), the neuronal nucleus size vs. neuronal cytoplasm size relationship (h), the relation between neuronal nucleus and neuronal cell (i), the neuronal nucleus size vs. the profile area of the space delineated by the satellite cells relationship (j), the profile area of the satellite cells vs. neuronal nucleus size relationship (k) and the profile area of the satellite cells vs. neuronal nucleus‐cytoplasm relationship. Please note that the numerical density of nerve cells (a,b) is split up into two graphs: age 0–20 and age 20–100. Also, note the different y‐axis in RA1‐3 (g‐i) compared with RA4 (j).

Perikaryon

The smallest neuronal profile area was found in fetal sections. There was a steep increase in profile area until the onset of puberty (r = 0.82; P < 0.00001; see Fig 3c,e). Afterwards, growth rate was reduced, thus no significant changes occurred till the age of 75. All in all, there was still a significant positive correlation in perikaryon profile area with increasing age (r = 0.62; P < 0.001; see Fig 3c,e).

Nucleus

The nuclear profile area was minimal in fetal sections. There was a statistically significant increase until the age of 4 years (r = 0.79, P < 0.00001; see Fig 3d). In higher age groups, there was a slight decrease in nuclear profile area. All in all, there was still a significant positive correlation with increasing age (r = 0.57; P < 0.00001; see Fig 3d).

Satellite cells

The lowest ratio of satellite cells per neuron occurred in fetuses and then increased steeply till puberty (r = 0.79; P < 0.00001; see Fig 3f). Thereafter, the ratio remained stable (average of six satellite cells per neuron). The ratio had a significant positive correlation coefficient throughout the whole age range (r = 68; P < 0.001, see Fig 3f).

Satellite cells–cytoplasm interspace

The lowest profile area of the satellite cells–cytoplasm interspace was present in fetuses; there was a steep increase in size until the age of 4 years (r = 0.70; P < 0.00001; not shown). There was a non‐significant reduction in profile area of the SC‐interspace in higher age groups but a significant positive correlation with increasing age (r = 0.65; P < 0.001; not shown).

RA1–RA4

In the following, we relate different compartments of a neuron to one another and compare the changes of these relationships depending on the patient's age. Relations of RA1 (nucleus/cytoplasm), RA2 (nucleus/cell), and RA3 (nucleus/ring) had a negative correlation with age (r RA1 = −0.69; r RA2 = −0.69; r RA3 = −0.67; P RA1, RA2,,RA3 < 0.00001, see Fig 3g‐i). This suggests that the nucleus‐cytoplasm relationship shifts with increasing age to a higher cytoplasm ratio, i.e. a relatively higher cytoplasm amount. The cytoplasm growth is also the main reason for the neuronal growth. This was proved by the development of RA4 (cytoplasm/cell): there was a highly significant positive correlation with age (r RA4 = 0.69; P RA4 < 0.00001, see Fig 3j).

SAT1–SAT2

In the following, we relate the number of satellite cells per neuron to different compartments of the neuron and compare the changes of these relationships with aging.

SAT1 (satellite cells/nucleus) increased continuously till the onset of puberty (r SAT1 = −0.48; P SAT1 ≤ 0.00001, see Fig 3k), but there were no overall dynamics with increasing age. SAT 2 (satellite cells/cytoplasm) decreased till the second year of life and then remained stable (r SAT2 = −0.46; P SAT2 = 0.00001, see Fig 3l).

Discussion

In the following paragraphs, we will compare our results with the results from other studies, both human and animal studies. At the beginning we will deal with macroscopic and quantitative changes, and finally with further histological observations, which have been described in other studies as age‐related changes (lymphocytic reaction, protoplasmatic stalks, fenestrations, binucleated neurons, neuro‐axonal dystrophies, glomerular dendrites, neurofibrils, reduction of Nissl substance, and storage of lipopigments).

Classification according to the appearance

The otic ganglion was classified differently over time: Tanaca (1932) described the ganglia as oval (n = 24), spindle‐shaped (n = 23), polygonal (n = 6), roundish (n = 4) or triangular (n = 3) (Tanaca, 1932). Roitman et al. specified the shape of the otic ganglion as ‘ganglionic’ (60%), that is, round to oval, or as ‘some thickening’ (13%). In 27% of the cases, he was not able to identify a distinct anatomical structure and registered them as ‘missing ganglia’ (Roitman et al. 1990).

We characterized the appearance of the specimens into the following three types: (i) compact type, (ii) lobulated type, and (iii) disperse type (see Fig 4). The compact type is macroscopically round to oval and of ‘ganglionic’ shape. In about half of the specimens, the otic ganglion does not have a compact appearance. This is mainly the result of how nerve bundles enter and leave the otic ganglion. Generally speaking, the higher the number of leaving branches, the less lobulated and the more disperse the ganglion appear. In addition, neuroblasts migrate at various paces along the mandibular nerve. If the neuroblast migration proceeds evenly, there will be a smaller degree of branching and the ganglion will appear more like a compact, ‘ganglionic’ or ‘classical’ type. Assignment of samples to each morphological type proved to be uncomplicated; however, certain individuals had otic ganglia belonging to different types; for example, in the same individual there was a compact ganglion on the right side and a disperse one on the left side. The distribution of the three ganglionic types was as follows: compact type in 51.7% of cases (89 specimens), lobulated type in 23.3% of cases (40 specimens), and disperse type in 25% of cases (43 specimens).

Figure 4.

Figure 4

Open in a new tab

Three types of appearance of the otic ganglion. (a) compact type; (b) lobulated type; (c) dispersed type. (d) A ganglion cell and the measured components. (a‐c) Bielschowsky staining, (d) HE ‐stain.

Roitmans et al.'s ‘ganglionic form’ corresponds approximately in amount and description to our compact type. The lobulated type can either be subsumed under the classical ‘ganglionic’ form or as ‘some thickening’; it can also be correlated with Tanaca's description of a plate‐like or disc‐like impression. The disperse type correlates most likely with Roitman et al. data of ‘missing ganglia’ (27% in their material) and maybe with Tanaca's ‘polygonal plate’ (10% in his material) (Tanaca, 1932; Roitman et al. 1990).

Size and volume

The indications of size fluctuate considerably. For example, Arnold defined the volume of the otic ganglion as 2 × 1.5 × 0.4 mm (length × height × thickness), Tanaca nearly an eightfold volume (4.9 × 2.4 × 0.8 mm) and Roitman et al. even a bit more (4 × 3 × 1.5 mm) (Arnold, 1829; Tanaca, 1932; Roitman et al. 1990).

There are many reasons for this deviation: the aforementioned differences in types (compact vs. disperse), examinations on different population groups (Europeans vs. Japanese), unavoidable artificial influences (shrinkage because of fixation), precise preparation of surrounding fat tissue, measurement errors or use of different measuring instruments, and so on. As exact measurements are nearly impossible and their usefulness is doubtful, we only specified the largest diameter of compact or lobulated types (the extension of the disperse type cannot be seen macroscopically). We report values between 2 and 4 mm; an increase in size was clearly evident within the first two decades of life.

Neural growth during the first years of life

Lasowsky (1930) reported that neurons in human heart ganglia in adults (parasympathetic intramural ganglia of the vagus nerve) have a size between 32 and 40 μm. He observed an increase in the size of these neurons during age: 18–20 μm in newborns compared with 45–50 μm in elderly people. Neurons in the ganglia of the sympathetic trunk seem to be smaller; Stochdorph (1961) described their diameter as ranging between 25 and 32 μm.

We observed the fastest growth rate in the first year of life and the biggest perikaryon size in adults (25–44 years, 38.64 ± 3.46 μm, see all data in Table 1).

Quantification of neurons during aging

Baker & Santer (1988) observed a significant reduction of neurons and numerical density in the celiac‐superior mesenteric ganglion (CSMG) in rats in their third quarter of life. They recognized the same process in superior cervical ganglion (SCG) in the last quarter of life of rats, so in comparison with CSMG, there was a little delay. An earlier neuronal loss was detected in the myenteric plexus Auerbach in rats: 40–60% loss of neurons in the second half of life (6–24 months). Unfortunately, those authors did not examine the otic ganglion.

Their curve progression of the SCG correlated best with our data, as a reduction in the numerical density in the otic ganglion occurred mostly in the last third of life (Baker & Santer, 1988).

Quantification of satellite cells during aging

We did not find detailed figures on the number of satellite cells in the literature. In our material most neurons in adults were surrounded by five to eight satellite cells (median: 5; mean value: 6.7; standard deviation: 0.7). This number only refers to satellite cells lying between capsule and perikaryon.

Lymphocytic reaction

In six patients there were lymphocytic infiltrates; four of them were over 60 years old. An interpretation of these results in view of aging processes is highly speculative. Lymphocytic reactions might be due to neuronal decay products and were therefore interpreted as a degenerative feature by Schmidt (1991).

Protoplasmatic stalks and fenestration

Hermann described protoplasmic stalks, which he found to occur for the first time in individuals during the third decade of life (Hermann, 1950). Results from our collection of samples demonstrated such structures already in young infants. Quantitative expression of these alterations was differing distinctly between individuals; there was no clear correlation with age.

Riquier also observed stalked protoplasm lobes frequently in human otic ganglia as well as ‘fenestration’ (Riquier, 1913). The latter phenomenon had also been described many times in animals (Slavich, 1932; Pines & Narowtschatowa, 1934). We can confirm both observations.

Hermann assumes that the causes of this finding are ‘degradation products by nerve cells forming anastomoses’ and ‘metabolites from growth processes’ (Hermann, 1950). Other references raise contradicting views, which suggests an age‐dependent increase of fenestration (Stöhr, 1957).

We first found fenestrated neurons in a 3‐year‐old infant but overall there was a tendency to increased occurrence with age.

Binucleated neurons

Very rarely we observed neurons with two nuclei. That may be because the otic ganglion is part of the parasympathetic system: Hermann reported that neurons with two nuclei are less frequent in parasympathetic ganglia than in sympathetic ganglia (Hermann, 1950). But it has also been demonstrated in animals that the total number of binucleated neurons decreases with age (Ehlers, 1951).

We would support the prevailing opinion that binucleated cells in ganglia are typical of physiological cellular proliferation. Some authors presented a ‘degenerative’ way of nuclear division, ‘degeneration by irritant agents’ (Botar, 1956; Stöhr, 1957). None of our findings matched the descriptions of these authors.

Neurofibrils

We can also confirm the presence of neurofibrils mentioned by Pines & Narowtschatowa (1934). These may also be a result of agglutinated neurofilaments that occur during the fixation process and can be made visible by silver stains.

Glomerular dendrites

Many authors described a more complex development of the dendritic tree, as the processes became more numerous, more voluminous, more hypertrophic, more ramified, and thus more entangled. The location of the branching spot changed with age: processes arise from one pole in newborns but more diffusely from the soma in elderly people (Hermann, 1950). These glomerulus‐like branches of processes (‘glomerular dendrites’) had been described in humans and oxen by Riquier (1913), but were missing in horses, sheep, cattle and apes (Müller & Dahl, 1910; Slavich, 1932; Pines & Narowtschatowa, 1934).

We already observed glomeruli dendriticiti in a 3‐year‐old child. They became more complex and more common in ascending age groups. However, we cannot concluded whether these formations are a consequence of aging, disease or a hypertrophic appearance in the course of physiological compensation.

Neuro‐axonal dystrophies (NADs)

Older papers described oval, elongated swellings of processes (‘sphere phenomenon’), which are called neuro‐axonal dystrophies (NADs). Mostly these lesions can be found near the perikaryon and dendrites; they appear as argyrophilic swellings or spheres at preterminal axons and synapses. The sphere‐like neuro‐axonal dystrophies are a typical and constant sign of aging, more frequently among men and also found in the CNS in various metabolic disorders (Schmidt, 1991). These neurons show different alterations in electron microscopy: swelling of mitochondria, storage of ‘residual bodies’ in axons, and irregularity and dilation of the endoplasmic reticulum and the Golgi apparatus (Santer et al. 1980).

We found NAD lesions for the first time in a 7‐month‐old child. A continuous increase was evident after the age of 50. They might be a sign of synaptic changes with increasing turnover and regeneration.

Nissl substance and lipopigments

We agree with Kunz that the Nissl substance reduces during aging – this phenomenon can be explained by an increase of chromatolysis in elderly people (Kunz, 1938). We have also seen storage of neuronal lipopigments such as lipofuscin and neuromelanin in older neurons, which is a well‐known occurrence in neurons of the autonomic nerve system (Hervonen et al. 1986).

Short conclusion

Our data show that the density of nerve cells decreases rapidly in the first 4 years of life due to the growth of the cytoplasm of the neurons, whereas the number of satellite cells increases till puberty. The parameters we measured remained relatively stable over the next decades of life. Even fetuses of a gestation age of 20 weeks had developed otic ganglia.

The first signs of neuronal degeneration could be found in teenagers but became clearly evident in elderly people (> 60 years). Degenerative effects are the main reason for the general decrease over time of the size of the otic ganglion.

Conflicts of interest

Authors declare no conflict of interests.

Ethics

Following the Austrian legal procedures, every patient who dies in a hospital can be autopsied for diagnostic and/or scientific purposes.

Acknowledgements

The examined samples of the otic ganglion were kindly supplied by the Department of Forensic Medicine, University of Kasachstan, Karaganda (Head: Prof. Dr A. N. Samoilitschenko). The material from the fetuses and newborns was provided by the Municipal Hospital for Gynecology and Obstetrics, Karaganda (Head: Dr med. T. L. Kim).

References

  1. Akbas M, Gunduz E, Sanli S, et al. (2016) Sphenopalatine ganglion pulsed radiofrequency treatment in patients suffering from chronic face and head pain. Braz J Anesthesiol 66, 50–54. [DOI] [PubMed] [Google Scholar]
  2. Anders K, Kautzky R (1955) Die Frühentwicklung der vegetativen Hals‐ und Kopfganglien des Menschen. Z Anat 119, 55–84. [PubMed] [Google Scholar]
  3. Arnold F (1829) Dissertatio inanguralis medica sistens observationes nonnulas neurologicas de partre cephalica nervi simpathici in homine. Z Physiol 2, 125–172. [Google Scholar]
  4. Baker D, Santer R (1988) Morphometric studies on pre‐ and paravertebral sympathetic neurons in the rat: changes with age. Mech Ageing Dev 42, 139–145. [DOI] [PubMed] [Google Scholar]
  5. Barloese MC, Jurgens TP, May A, et al. (2016) Cluster headache attack remission with sphenopalatine ganglion stimulation: experiences in chronic cluster headache patients through 24 months. J Headache Pain 17, 67. [DOI] [PMC free article] [PubMed] [Google Scholar]
  6. Botar J (1956) Qualitative und quantitative Untersuchung der Nervenzellen des Ganglion coeliacum im Alter. Alterserscheinungen der Sympathischen Nervenzellen. Acta Anat 26, 192–245. [PubMed] [Google Scholar]
  7. Edvinsson L, Hara H, Uddman R (1989) Retrograde tracing of nerve fibers to the rat middle cerebral artery with true blue: colocalization with different peptides. J Cereb Blood Flow Metab 9, 212–218. [DOI] [PubMed] [Google Scholar]
  8. Edvinsson L, Tajti J, Szalardy L, et al. (2018) PACAP and its role in primary headaches. J Headache Pain 19, 21. [DOI] [PMC free article] [PubMed] [Google Scholar]
  9. Ehlers P (1951) Über Altersveränderungen an Grenzstrangganglien von Meerschweinchen. Anat Anz 81, 24–34. [PubMed] [Google Scholar]
  10. Gibbins I (1990) Target related patterns of co‐existence of neuropeptide Y, vasoactive intestinal peptide, enkephalin and substance P in cranial parasympathetic neurons innervating the facial skin and exocrine glands of guinea pigs. Neuroscience 38, 541–560. [DOI] [PubMed] [Google Scholar]
  11. Grunditz T, Uddman R, Sundler F (1994) Origin and peptide content of nerve fibers in the nasal mucosa of rats. Anat Embryol 189, 327–337. [DOI] [PubMed] [Google Scholar]
  12. Gundersen H (1977) Notes on the estimation of the numerical density of arbitrary profiles: the edge effects. J Microsc 111, 219–223. [Google Scholar]
  13. Hermann H (1950) Mikroskopische Beobachtungen über Altersveränderungen an den menschlichen Herzganglien. Zschr Anat Entwgesch 114, 685–719. [Google Scholar]
  14. Hervonen A, Helen P, Partanen M, et al. (1986) The sympathetic neuron as a model of neuronal aging In: Neurohistochemistry: Modern Methods and Applications (eds. Panula PPH, Soinila S, Eränkö O.). New York: Liss. [Google Scholar]
  15. Izumi H, Karita K (1992) Somatosensory stimulation causes autonomic vasodilatation in cat lip. J Physiol 450, 191–202. [DOI] [PMC free article] [PubMed] [Google Scholar]
  16. Kaji A, Shigematsu H, Fujita K, et al. (1988) Parasympathetic innervation of cutaneous blood vessels by vasoactive intestinal polypeptide immunoreactive and acetylcholinesterase‐positive nerves: histochemical and experimental study on rat lower lip. Neuroscience 25, 353–362. [DOI] [PubMed] [Google Scholar]
  17. Kuchiiwa S, Izumi H, Karita K, et al. (1992) Origins of parasympathetic postganglionic vasodilator fibers supplying the lips and gingivae, an WGA HRP study in the cat. Neurosci Lett 142, 237–240. [DOI] [PubMed] [Google Scholar]
  18. Kunz A (1938) Histological variations in autonomic ganglia and ganglion cells associated with age and disease. Am J Pathol 14, 783–795. [PMC free article] [PubMed] [Google Scholar]
  19. Lainez MJ, Marti AS (2016) Sphenopalatine ganglion stimulation in cluster headache and other types of headache. Cephalalgia 36, 1149–1155. [DOI] [PubMed] [Google Scholar]
  20. Lasowsky J (1930) Normale und pathologische Histologie der Herzganglien des Menschen. Virchows Arch Path Anat 279, 464–465. [Google Scholar]
  21. Leblanc G, Trimmer B, Landis S (1987) Neuropeptide Y like immunoreactivity in rat cranial parasympathetic neurons: coexistence with vasoactive intestinal peptide and choline acetyl‐transferase. Proc Natl Acad Sci USA 84, 3511–3515. [DOI] [PMC free article] [PubMed] [Google Scholar]
  22. Lovasova K, Sulla IJ, Bolekova A, et al. (2013) Anatomical study of the roots of cranial parasympathetic ganglia: a contribution to medical education. Ann Anat 195, 205–211. [DOI] [PubMed] [Google Scholar]
  23. Müller R, Dahl W (1910) Die Beteiligung des sympathischen Nervensystem an der Kopfinnervation. Dtsch Arch Klin Med 1–2, 48–52. [Google Scholar]
  24. Narouze S (2014) Neurostimulation at pterygopalatine fossa for cluster headaches and cerebrovascular disorders. Curr Pain Headache Rep 18, 432. [DOI] [PubMed] [Google Scholar]
  25. Pines L, Narowtschatowa K (1934) Über die Morphologie des Ganglion oticum. Z Zellforsch Mikrosk Anat 20, 746–778. [Google Scholar]
  26. Retzius G (1880) Untersuchungen über die Nervenzellen der zerebrospinalen Ganglien und der übrigen peripherischen Kopfganglien. Archiv Anat und Physiol, Anat Abt 1, 369–398. [Google Scholar]
  27. Riquier C (1913) Sulla fine struttura del ganglio otico. Riv Pat Nerv 1913, 18–22. [Google Scholar]
  28. Roitman R, Talmi Y, Finkelstein Y, et al. (1990) Anatomic study of the otic ganglion in humans. Head Neck 12, 503–506. [DOI] [PubMed] [Google Scholar]
  29. Ruskell G (1993) Distribution of otic postganglionic and recurrent mandibular nerve fibres to the cavernous sinus plexus in monkeys. J Anat 182, 187–195. [PMC free article] [PubMed] [Google Scholar]
  30. Sangiacomo C (1969) Submicroscopic organization of the otic ganglion of the adult rabbit. Z Zellforsch Mikrosk Anat 95, 290–309. [DOI] [PubMed] [Google Scholar]
  31. Santer R, Partanen M, Hervonen A (1980) Glyoxylic acid fluorescence and ultrastructural studies of neurones in the coelac superior mesenteric ganglion of the aged rat. Cell Tissue Res 211, 475–485. [DOI] [PubMed] [Google Scholar]
  32. Schmidt R (1991) Pathology of the sympathetic nervous system In: The Pathology of the Aging Human Nervous System (ed. Duckett SD.). Philadelphia: Lea & Febiger. [Google Scholar]
  33. Schoenen J (2015) Sphenopalatine ganglion stimulation in neurovascular headaches. Prog Neurol Surg 29, 106–116. [DOI] [PubMed] [Google Scholar]
  34. Segade L, Suarez Quintanilla D (1988) Otic ganglion parasympathetic neurons innervate the pulp of the mandibular incisor of the guinea pig. Neurosci Lett 90, 33–38. [DOI] [PubMed] [Google Scholar]
  35. Senger M, Stoffels HJ, Angelov DN (2014) Topography, syntopy and morphology of the human otic ganglion: a cadaver study. Ann Anat 196, 327–335. [DOI] [PubMed] [Google Scholar]
  36. Sharkey K, Templeton D (1984) Substance P in the rat parotid gland: evidence for a dual origin from the otic and trigeminal ganglia. Brain Res 304, 392–396. [DOI] [PubMed] [Google Scholar]
  37. Shimizu T (1994) Distribution and pathway of the cerebrovascular nerve fibers from the otic ganglion in the rat: anterograde tracing study. J Auton Nerv Syst 49, 47–54. [DOI] [PubMed] [Google Scholar]
  38. Siessere S, Vitti M, Sousa LG, et al. (2008) Anatomic variation of cranial parasympathetic ganglia. Braz Oral Res 22, 101–105. [DOI] [PubMed] [Google Scholar]
  39. Slavich E (1932) Confronti fra la morfologia di gangli del parasympathico encefalico e del simpatico cervicale con speciale riguardo alla struttura del ganglio ciliare. Z Zellforsch Mikrosk Anat 4, 65–77. [Google Scholar]
  40. Stochdorph O (1961) Normale und pathologische Anatomie des vegetativen Nervensystems In: Lehrbuch der Speziellen Pathologischen Anatomie (ed. Staemmler M.). Berlin: Walter de Gruyter. [Google Scholar]
  41. Stöhr P (1957) Mikroskopische Anatomie des vegetativen Nervensystems In: Handbuch der Mikroskopischen Anatomie (ed. Bargmann W.). Berlin: Springer. [Google Scholar]
  42. Suzuki N, Hardebo J, Owman C (1988) Origins and pathways of cerebrovascular vasoactive intestinal polypeptide positive nerves in rat. J Cereb Blood Flow Metab 8, 697–712. [DOI] [PubMed] [Google Scholar]
  43. Suzuki N, Hardebo J, Owman C (1990) Origins and pathways of choline acetyltransferase positive parasympathetic nerve fibers to cerebral vessels in rat. J Cereb Blood Flow Metab 10, 399–408. [DOI] [PubMed] [Google Scholar]
  44. Tanaca T (1932) Ganglion oticum. In: Arbeiten der 3. Abteilung des anatomischen Institutes Kyoto. Vol. 3, pp. 41–49. Kyoto, Japan: Abteilung des anatomischen Institutes. [Google Scholar]
  45. Uddman R, Edvinsson L (1989) Neuropeptides in the cerebral circulation. Cerebrovasc Brain Metab Rev 1, 230–252. [PubMed] [Google Scholar]
  46. Uddman R, Edvinsson L, Hara H (1989) Axonal tracing of autonomic nerve fibers to the superficial temporal artery in the rat. Cell Tissue Res 256, 559–565. [DOI] [PubMed] [Google Scholar]
  47. Weis S (1991) Morphometry in the neurosciences In: Image Processing and Computer Graphics. Theory and Applications. OCG Schriftenreihe. (eds. Wenger E, Dimitrov L.), Oldenburg: Verlag. [Google Scholar]

Articles from Journal of Anatomy are provided here courtesy of Anatomical Society of Great Britain and Ireland

RESOURCES