An 8.7 mL venous blood sample was drawn under aseptic conditions. The blood was aspirated with a 21 G needle into a 10 mL syringe preloaded with 1.3 mL of anticoagulant citrate dextrose (ACD) solution (in rats: 2 mL venous blood sample mixed with 0.3 mL of ACD). Each blood sample was centrifuged for 15 minutes at 3000 RPM, 72 g, at 4℃, resulting in the following three layers: an inferior layer composed of red cells, an intermediate layer composed of white cells, and a superior layer made up of plasma. The 6 mL plasma layer was centrifuged for another 5 minutes at 1000 g in order to obtain a two-part plasma sample: the upper part consisting of 5.5 mL of platelet-poor plasma (PPP) and the lower part consisting of 0.5 mL of platelet-rich plasma (PRP).22 The PPP was then gently aspirated with a pipette and placed in a sterile injection bottle, being careful not to mix the PPP with PRP.

The injection bottle was attached to a dental syringe and heated at 100℃ for 12 minutes in a heating machine (ALSA S-1, Genexel-Seine, Seoul, Korea). The injectable plasma gel was then ready for use (Fig. 1). An injection needle was attached to the dental syringe, and the plasma gel was injected into the target site.

The present animal study was performed after obtaining approval from the Animal Research Committee, Gyeongsang National University. Thirty Sprague-Dawley rats (Samtako, Osan, Korea) with body weights of over 200 grams were used in the study.

For preparation of the plasma gel, blood was withdrawn from the tail vein after the rats were anesthetized with i.p. injections of xylazine hydrochloride (10 mg/kg) (Rompun, Bayer, Germany) and Zoletile 50 (10 mg/kg) (Virbac Laboratories, Carros, France). The acquired blood was processed to produce plasma gel following the above protocol. Two 0.2 mL subcutaneous injections of plasma gel were given to each animal at intervals of 5 mm along the dorsal area. Animals were examined at weekly to monthly intervals for signs of irritation, ulceration, or inflammation, and the amount of plasma gel remaining in situ was measured by electric measurement.

The skin and underlying tissue at the injection sites were removed for analysis at 24 hours (n=3), 30 days (n=3), 90 days (n=3), and 180 days (n=21) post injection. The tissues were dehydrated and processed according to standard paraffin-embedding procedures. Five-micrometer sections were stained with hematoxylin and eosin (H&E) for evaluation of histological changes after injection of plasma gel.

The H&E stained sections were examined for inflammatory cell infiltration according to time of absorption. Inflammatory cells in the subcutaneous space were counted at ×100 magnification from two representative areas (superior and inferior injection sites) for each of the animals. The counts were averaged and used for comparisons.

The present clinical study was conducted in patients diagnosed with unilateral vocal cord palsy in the Department of Otolaryngology at Gyeongsang National University Hospital, Korea, between January 2010 and March 2011. Eleven patients (8 males and 3 females) who wished to relieve their voice and aspiration symptoms were enrolled. The mean age of the patients was 53 years (range, 25-80 years). The inclusion criteria were as follows: patients having dysphonia associated with idiopathic unilateral vocal fold palsy (VFS), 4 cases; and patients having mobile vocal folds with glottal insufficiency due to either chest surgery or thyroid surgery, 7 cases. This study was performed after obtaining approval from the Institutional Review Board of Gyeongsang National University Hospital, and all patients provided signed written consent.

Percutaneous injection of plasma gel was performed under local anesthesia by one author (SHW) using the thyrohyoid membrane approach. Through the midline area of the thyroid notch, by penetrating the thyrohyoid membrane, the tip of the needle was exposed at the upper vocal cord. While observing the vocal cord of the patient using a flexible laryngoscope, the prepared plasma gel was injected to the muscle layer of the vocal cord. The injection volume was approximately 1.0 mL. After the procedure, the patient rested in a ward for 2-3 hours, and was discharged on the same day.

The standardized voice handicap index (VHI) before and at the fourth week after injection was used. Three domains (functional, physical, and emotional) are included in the index, and each contains 10 questions requiring patients to indicate how frequently they experienced each situation. Responses to each question were graded from 0 to 4, depending on the perceived degree of handicap.

In addition, a transoral video-stroboscopic examination was performed in all patients before and at 1, 2, and 3 months after injection for structural evaluation. The findings were analyzed and sorted into three categories (disappeared, improved, or no change) by comparing the size of the vocal gap at the follow-up visit with that before the procedure. Acoustic recordings and phonatory function studies were also performed on patients. A voice sample of the patients uttering a sustained vowel sound ("aaa") for 3 seconds at a conversational pitch and loudness was acquired. All voice inputs were recorded and sampled with a multidimensional voice program (MDVP) (Model 4500; Kay Elemetrics Corp., Lincoin Park, NJ, USA) for voice analysis. Four MDVP parameters were used in this study: average fundamental frequency (F0), jitter (jitt %), shimmer (shim %), and noise to harmony ratio. The aerodynamic study was conducted with a computerized system (Phonatory Aerodynamic System, Model 6600; Kay Elemetrics Corp.). The maximum phonation time (MPT) during vowel emission was expressed in seconds.

Data were analyzed using SPSS software, version 15.0 (SPSS Inc., Chicago, IL, USA). The acoustic analysis data before and after injection were compared using a paired Student's t-test. Statistically significant differences in VHI before and after injection were determined with the Wilcoxon signed rank test. Null hypotheses of no difference were rejected for p-values less than 0.05.

The plasma gel was a semi-solid and easily injectable material which was stable at the injection site (Fig. 1E). The plasma gel was maintained well at the same consistency for over 1 year in a sealed injection bottle; however, when exposed to room air (opened bottle), the plasma gel disappeared within one month (Fig. 1D and F). Under an electron microscope, plasma gel was seen to consist of many same-sized round particles, which represented a consistent and homogenous protein feature (Fig. 1G).

Injection sites in the animals were first graded at 24 hours after injection, and the mean diameter was 13.07±0.4 mm (Fig. 2). The mean diameter was 11.8±0.9 mm at day 30, 8.5±1.35 mm at day 90, and 0.5±0.48 mm at day 180. We found that two (3.3%) injection sites eventually exhibited ulceration and scarring. These rare adverse events, typically occurring 1 month post injection, may be attributed to animal behavior (scratching and biting).

At 1 day post injection, there was no evidence of infiltration of acute inflammatory cells (Fig. 2). At day 30, slight inflammatory cellular infiltration was noted, and a mature fibrous capsule surrounded the plasma gel depot at day 90. Deep dermal cysts, which were lined by flattened epithelial cells, were prominent by 3 months at injection sites in the animals. However, chronic inflammation and foreign body giant cells around the plasma gel depot were absent. At day 180, there was no visible fibrous capsule or cell infiltration.

Eleven patients were enrolled in this study; all received a percutaneous plasma gel injection at a site of vocal cord palsy via the thyrohyoid membrane under topical anesthesia. The average volume of injection material was 1.2 mL per vocal fold.

The stroboscopic findings showed that the vocal gap disappeared and the grade of hoarseness improved in all patients (Table 1). The preoperative and postoperative acoustic and aerodynamic parameters also improved significantly. A clear and significant improvement was visible for the mean values of jitter % (p<0.001), shimmer % (p<0.001), and MPT (p<0.05). Mean VHI score also showed significant improvement for all domains after injection (p<0.001) (Fig. 3).

Among the video stroboscope findings, vocal cord volume began to decrease gradually after 2 months in most patients. In 3 out of 11 patients, at 2 months after injection, voluntary mobility of the vocal cord was observed (Table 2). Meanwhile, in 6 of the 11 patients, hoarseness redeveloped gradually, and voluntary mobility of the vocal cord was absent, thus after 3 months, other artificial materials were re-injected. In the other 2 cases (cases 1 and 9), although vocal fold palsy did not recover, the patients' symptoms were not severe, and thus the patients were maintained under follow-up observation without undergoing additional treatments. The only complication in this study was a hematoma in the area injected with plasma gel in one patient who was taking anticoagulant agents. However, the hematoma was small and there was no symptom of respiration, so we continued to observe the hematoma in the outpatient department. One month later the hematoma disappeared. In the other patients, no complications related to plasma gel injection were observed.