The Presence of Mechanoreceptors and Cortical Connections Support a Proprioceptive Role for Knee Plicae

Purpose

To determine histologically and electrophysiologically whether the plicae have a proprioceptive function.

Methods

Tissue samples were obtained from 36 plicae in 29 knees (including 7 bilateral cases). The samples included 20 suprapatellar, 10 medial patellar, and 6 infrapatellar plicae. Kim and Choe’s classification was used to describe the patterns of the plicae.

Results

Twenty-nine patients who underwent knee arthroscopy were included in the study. Of these, 14 knees were on the left side and 15 on the right. The patients’ ages ranged from 12 to 78 years (mean age, 42 years). Golgi tendon organs (type III) were the most frequently observed mechanoreceptors. No free nerve endings (type IV) were identified in any of the plicae. Electrical stimulation of 4 plicae and 2 anterior cruciate ligaments was performed, and accurate, expected responses were obtained.

Conclusions

Histologic and neurophysiologic findings have shown the presence of mechanoreceptors in the knee plicae. Following the identification of these mechanoreceptors, the establishment of a neural connection between the plicae and the cerebral cortex suggests that the plicae are functional structures involved in proprioception. However, free nerve endings were evaluated in limited numbers and could not be shown.

Clinical Relevance

Plicae in the knee are considered rudimentary structures with the potential to become pathological. However, the plicae may not be useless structures but may improve the proprioceptive ability of the knee.

Synovial plicae are considered the remnants of the septa that divide the joint into 3 compartments during embryonic development of the knee. These compartments converge into a single cavity at the 12th week of fetal growth. If the reabsorption of the divisions between the spaces fails or is inadequate, a plica can occur. There are 4 plicae in the knee: the suprapatellar plica (SP), the infrapatellar plica (IP) (ligamentum mucosum), the medial patellar plica (MP) (shelf), and the lateral patellar plica. The medial and lateral patellar plicae may not be remnants of the septa of the prosthetic compartments; rather, they appear to be remnants of mesenchymal tissue related to developmental conditions. Reabsorption of the suprapatellar septum may continue in the postpartum period. Most plicae can be found in the knee and have been seen on arthroscopies, but most remain physiologic and should not be seen as potentially pathologic. ^, Sometimes they become pathological or symptomatic and then lead to a disorder called plica syndrome. ^, It has been stated that plicae can cause symptoms, but the syndrome is overdiagnosed; normal synovial plicae can be removed. ^,

Proprioception is considered a specific variation of the sense of touch; it includes the senses of joint movement (kinesthesia) and joint position (joint position sense). Proprioception can have an important role in dynamic joint stability, acute and chronic injuries, pathological wear, and rehabilitation. In general, histologic, neurophysiologic, and clinical features have been looked at when assessing proprioceptive abilities. ^, Four types of articular ends have been identified: type I (Ruffini), type II (Pacini), type III (Golgi), and type IV (free nerve endings); their functions include chemoreception and pain reception. ^,^, In the human knee, mechanoreceptors have been documented in the anterior cruciate ligament (ACL), posterior cruciate ligament (PCL), meniscus, and lateral collateral ligament. ^,^,

Somatosensory evoked potentials (SEPs) measure electrical potentials evoked in the cerebral cortex upon stimulation of a peripheral neuroreceptor. Thus, the proprioceptive functions of a particular structure have been proven by the use of SEPs. By this method, the presence of active mechanoreceptors were demonstrated in the intact ACL. ^,

The aim of this study was to determine histologically and electrophysiologically whether the plicae have a proprioceptive function. We hypothesize that plica tissue contains mechanoreceptors, which have been associated with knee proprioception.

Methods

Patients who underwent arthroscopic plica excision between 2002 and 2006 were evaluated retrospectively. Informed consent was obtained from all patients who were included in the study, and ethics committee permission was obtained. Parental consent was also obtained from 4 patients under 18 years of age (patients 10, 18, 19, 20).

The study included patients aged between 12 and 78 years who underwent arthroscopy for various intraknee pathologies and accepted to participate. Patients with previous knee surgery, connective tissue disease, thin plica, and high probability of capsule involvement in the sample were excluded. Some cases were also excluded from the study if the plica was not thick enough and the capsule tissue was inadvertently removed during the biopsy procedure and if there was a risk of misinterpretation.

The presence of plica was detected by clinical examination and imaging methods combined, and then arthroscopy was performed. Kim and Choe’s classification was used to describe the patterns of the plicae. Biopsy specimens were obtained from a minimum of 3 sites during arthroscopy with the help of probe, cautery, and biting and plucking instruments. Small and very thin plicae were not biopsied. Due to the similarity of capsular receptors to mechanoreceptors, biopsy specimens were taken from more central areas of the plicae.

Since electrophysiological studies were to be performed in patients under general anesthesia, another informed consent was obtained from 3 patients who accepted being operated on under general anesthesia. Electrophysiological study was performed in these patients under general anesthesia. Three SPs (2 medial types, 1 complete type) and 1 IP (separate type) were electrically stimulated to search for cortical evoked potentials.

Histologic Evaluation

Samples were evaluated histologically with 2 types of immunohistochemical materials. Twenty-two specimens from the first 17 cases underwent hematoxylin and eosin staining, Holmes histochemistry, and neurofilament protein (NFP) immunohistochemistry (Streptavidine-Biotin 2 System; Peroxidase) methods. Twelve SPs (4 medial, 3 complete, 5 arch types), 6 MPs (shelf type), and 4 IPs (separate type) were evaluated.

For the rest of the samples (14 plicae from 12 knees), the mechanoreceptors were determined by gold-chloride staining, hematoxylin and eosin, and NFP immunohistochemical tests (except for 1). Due to suboptimal staining by the O’Connor and Gonzales gold-chloride technique, a modification was used. Briefly, in this staining, the tissue was dehydrated with sucrose, treated with lemon juice solution, and impregnated with gold chloride, and the gold chloride was reduced. For each of these steps, tissue samples were fully immersed in 50 mL of the solution. The incubator not only allowed for temperature control but also provided a dark environment. Stirring and controlled temperature were kept constant for each step of the staining process, except for sucrose dehydration. The procedure was performed at both 20°C to 22°C and 30°C.

Evaluation of SEPs

The procedure was performed on 3 patients after additional informed consent was obtained. Three SPs (2 medial types, 1 complete type) and 1 IP in 3 knees were electrically stimulated to evaluate cortical evoked potentials. These plicae, except for 1, were also studied histologically.

Before the induction of general anesthesia, the posterior tibial nerve of the contralateral extremity was found by electrical stimulation; the nerve of each patient was first stimulated by a pair of 6-mm bipolar disc electrodes placed behind the medial malleolus over the posterior tibial nerve with the cathode proximal. A square wave pulse of 12 to 30 mA of a 0.2-ms duration was delivered at a rate of 2 per second.

Glysine irrigation fluid was used during arthroscopy due to its nonconductivity to electrical current. During arthroscopy, 4 plicae and 2 ACLs were stimulated by an epidural electrode introduced through a disposable cannula. A Teflon-coated monopolar electrode could not be used because our monitoring system was only compatible with an epidural electrode ( Fig 1 ). It has been shown that electrophysiological tracings obtained with epidural electrodes can be used instead of Teflon-coated electrodes by averaging.

Image, AltText currently not available — Fig 1

Cortical responses were monitored by surface electrodes secured to the scalp using the international 10/20 encephalography system; the reference electrode was placed at the C2 position, and the recording electrodes were placed at the C2 to C3 positions. ^, All responses were recorded over a bandwidth of 10 to 2,000 Hz for a duration of 100 ms, and 250 epochs were averaged to obtain 1 reading.

The cortical evoked responses are complex waveforms representing the sensory impulse as it travels through the sensory pathways to the cortex. Waves are labeled by polarity (P: positive, N: negative) and latency (expressed in ms). N1, P1, N2, and P2 latencies were obtained from recordings. Latency is the time measured between the stimulus of the structure and the cortical recording of the evoked potential; latencies of interest for lower extremities are P1 and N1.

Results

In 29 patients, 4 ACL injuries, 2 lateral meniscal cysts, 1 free body, 6 cartilage lesions, 5 cases of synovitis, 17 medial meniscal lesions, 7 lateral meniscal lesions, and 2 hypertrophic Hoffa lesions were found.

The mean age of the patients was 42.0 (range, 12-78) years. In 7 knees, biopsies were performed on the bilateral knee. Subtypes of plicae are shown in Table 1 . Arthroscopy was performed on 29 patients, and a total of 36 plica samples were taken from both knees of some of the patients. A total of 29 patients were evaluated, 14 from the left and 15 from the right knees. As seen in Table 2 , 18 were male and 11 were female.

Table 1

Demographic Data and Type and Subtypes of Plicae According to the Sung-Jae Kim Classification

Patient No.	Age, y	Sex	Type of Plicae	Patient No.	Age, y	Sex	Type of Plicae
1	28	M	SP (Medial type)	16	31	M	MP (Shelf type)
2	28	M	SP (Complete type) IP (Separate type)	17	53	F	SP (Arch type) MP (Shelf type)
3	25	M	M (Shelf type)	18	12	F	SP (Complete type)
4	41	M	IP (Separate type)	19	15	F	SP (Medial type)
5	44	F	SP (Medial type) MP (Shelf type)	20	14	F	SP (Medial type)
6	30	E	SP (Arch type)	21	37	M	IP (Separate type)
7	78	F	SP (Complete type)	22	55	M	SP (Complete type) MP (Triplicate type)
8	40	M	SP (Medial type) IP (Separate type)	23	29	M	SP (Lateral type)
9	64	F	SP (Arch type)	24	73	M	SP (Complete type)
10	17	M	SP (Medial type)	25	59	M	MP (Shelf type)
11	39	F	MP (Shelf type) IP (Separate type)	26	30	M	SP (Complete type)
12	60	F	SP (Complete type)	27	34	M	SP (Medial type) MP (Shelf type)
13	45	F	SP (Arch type)	28	62	M	MP (Shelf type)
14	42	M	SP (Arch type)	29	62	F	SP (Arch type)
15	70	M	MP (Shelf type)

NOTE. Neurofilament protein and hematoxylin and eosin staining methods were used.

IP, infrapatellar plica; MP, medial patellar plica; SP, suprapatellar plica.

Table 2

Latencies (N ₁ and P ₁) and Potentials (Μμ) to Cortical Evoked Potentials From Stimulation of the Plicae

	SP	MP	IP	Total
Medial type	7	—	—	7
Complete type	7	—	—	7
Arch type	5	—	—	5
Lateral type	1	—	—	1
Shelf type	—	9	—	9
Triplicate type	—	1	—	—
Separate type	—	—	5	5
Fenestra type	—	—	1	1
Total	20	10	6	36