Abstract
Our study aims to ascertain the results of speech therapy and surgery in patients who desire pitch alteration. Typically, patients desirous of an increase in their vocal pitch are male-to-female transpersons, and patients desirous of a drop in their pitch are puberphonia patients. This is a 3-year retrospective study of patients who have been operated for pitch alteration. This is a study of seven patients each of Wendlers glottoplasty (WG) and relaxation laryngeal framework surgery (RLFS). The records were reviewed for the preoperative and the 1-year post-operative fundamental frequency (FO), patient subjective satisfaction score (PSSS), and healing time. A total of 7/35 patients of puberphonia underwent RLFS in which the average preoperative F0 was 216 Hz and 1-year post-operative F0 was 114 Hz (p 0.004). The average pre- and post-operative PSSS was 4 and 9 (p 0.0004). Patients of WG had an average preoperative F0 of 153 Hz and 1-year post-operative F0 of 223 Hz (p 0.0005). The average pre and 1-year postoperative PSSS was 4 and 8 (p 0.002). The average healing time was 3 weeks 4 days in the RLFS patients and 8 weeks 5 days for WG patients. The results of RFLS yielded a high PSSS with a pitch drop of more than 100 Hz. The results of WG yielded an increase in the pitch of 70 Hz with a good PSSS. The average healing time taken following WG was double the time taken for RFLS.
Keywords: Puberphonia, Male to female transpersons, Relaxation laryngeal framework surgery, Type 3 thyroplasty, Wendler glottoplasty, Speech therapy
Introduction
Our study aims to ascertain the results of speech therapy and surgery as a combination or independent variable in patients who desire a pitch alteration. Typically, patients desirous of an increase in their vocal pitch are male-to-female (MTF) transpersons, and patients who desire a drop in their pitch are puberphonia patients unresponsive to speech therapy.
Voice modulation and speech therapy (ST) is offered to all patients desirous of a pitch change as a first-line management protocol. However, MTF transpersons who reach out to a voice clinic are often desirous of opting for a surgical option primarily. The reason may be that the individual cannot sustain a fundamental frequency (F0) without fatigue or having an occasional drop-in pitch during laughter, coughing, yawning, screaming, or throat-clearing [1].
However, in the puberphonia group of patients, the response to ST is extremely positive in most, and thus, this group needs to comply with the same before being considered for surgical intervention. The female-to-male transpersons typically do not need any surgery or even voice modulation as the hormonal therapy given to this group, i.e., testosterone, causes a thickening of the vocal folds which results in a drop in pitch over 6–9 months.
Our study aims to evaluate the 1 year post operative vocal outcomes of pitch change surgery, specifically Wendlers glottoplasty (WG) for pitch increase in MTF transpersons, and relaxation laryngeal framework surgery (RLFS) for pitch decrease in puberphonia patients.
Materials and Methods
This is a 3-year retrospective study where the records of our seven most recent MTF transpersons treated by WG (group A) and seven patients of puberphonia treated by relaxation laryngeal framework surgery (group B) were reviewed. The data was entered using MS Excel 2007 and analyzed using SPSS-16 software. Descriptive analysis for numerical data was summarized as mean with standard deviation (SD) and the P- value less than 0.05 was taken as statistically significant which was calculated using a student’s unpaired t-test. The preoperative and 1-year post-operative fundamental frequency along with the patient subjective satisfaction score (PSSS) were compared. The patients were asked to score their satisfaction with their voice on a scale of 1–10, and this scoring system was used to rate the voice, both pre-operatively and post-operatively. The healing or recovery time was also calculated, which was considered as the total time duration the patient needed speech therapy. The first author performed all surgeries, and the second author managed the patients regarding speech therapy.
WG surgery increases the F0 as the vibrating vocal fold length is decreased by almost half, by the creation of an anterior glottic web. All WG patients were operated under general anesthesia with laser, [Lumenis Acupulse CO2 (Carbon dioxide) laser system] used for freshening the anterior half of the medial edge of the membranous vocal fold (Fig. 1a). A total of 2 to 4 vicryl (6–0) sutures were taken including the vocal ligament in depth (Fig. 1b). Postoperatively, patients were advised two weeks of complete voice rest after which flexible laryngoscopy [Olympus flexible video laryngoscope ENF-VH] was performed to assess healing before initiating speech therapy. ST for transpersons is beneficial for both the segmental and non-segmental aspects of the speech. Pitch range exercises along with exercises to modify the resonance, quality, loudness, and intonation pattern were advised. The use of these techniques in real-life situations was also targeted.
Fig. 1.
a Intraoperative micro laryngeal image showing freshening of the anterior half of the medial edge of the membranous vocal fold using a CO2 laser. b 2 vicryl sutures were taken including the vocal ligament in depth
All RLFS patients were operated upon if they did not achieve sustained results despite a satisfactory ST trial focusing on pitch-lowering strategies such as the use of digital manipulation, glottal fry, and cough phonation. For patients younger than 25 years a minimum of 6 month period of ST was given and for patients 25 years and older, a minimum of 3 month period of ST was given. Surgery was performed under local anesthesia to allow for vocal feedback during the procedure. After exposing the anterior 8–9 mm of both thyroid alae a full-thickness vertical incision was made in the thyroid cartilage, 4–6 mm from the midline on either side. The inner perichondrium was left undisturbed. The detached 8–12 mm vertical midline cartilage strip was then gently pushed within the larynx, effectively relaxing the tension on the true vocal folds (Fig. 2a and 2b). A single 1–0 permanent suture was used to make a vertical mattress suture, which was very loosely anchoring the new edges of the two thyroid alae to keep the medial strip of the thyroid cartilage in a retruded position permanently. All patients were advised a week of voice rest followed by voice therapy. Post-operative voice therapy in puberphonia patients, focused on strengthening exercises to improve the range and stability of the vocal pitch.
Fig. 2.
a Pre-operative Stroboscopic picture of a puberphonia patient depicting tensed vocal folds. b Post-operative Stroboscopic picture of the same patient postoperatively showing relaxed vocal folds.
Results
In WG patients the average age was 24 years. The preoperative F0 was 153 Hz and the average 1-year post-operative F0 was 223 Hz which was statistically significant (p = 0.0005). The average preoperative PSSS was 4 and the average 1-year post-operative PSSS was 8 (p = 0.002).
The total number of puberphonia patients that visited our voice clinic during the study duration was thirty-five, of whom twenty-eight responded to ST within an average period of six weeks. The remaining seven patients who did not respond to ST despite good compliance were then counseled for RLFS, also referred to as type 3 thyroplasty. The average duration of the therapy trial before surgery was 13 weeks. In puberphonia patients, the average age of operated patients was 32 years. The preoperative F0 was 216 Hz and the average 1-year post-operative F0 was 114 Hz (p = 0.004). The average preoperative PSSS was 4 and the 1-year post-operative PSSS was 9 (p = 0.0004).
The healing time was 8 weeks and 5 days on average for WG patients 3 weeks and 4 days for RLFS patients. The average age, preoperative and 1-year post-operative fundamental frequency, and PSSS are tabulated in Table 1 for WG patients and in Table 2 for RLFS patients. The p-value and mean with SD for the pitch increase and PSSS are tabulated in Table 3 and for pitch decrease in Table 4.
Table 1.
Patient subjective satisfaction score (PSSS) in patients post Wendler Glottoplasty
| S. No | Age | Pre-op fundamental frequency | Post-op fundamental frequency | Pre-op PSSS | Post-op PSSS | Recovery time |
|---|---|---|---|---|---|---|
| 1 | 23 | 160 | 190 | 4 | 7 | 6w |
| 2 | 25 | 141 | 190 | 4 | 9 | 7w |
| 3 | 27 | 160 | 228 | 4 | 5 | 8w |
| 4 | 16 | 150 | 199 | 3 | 8 | 9w |
| 5 | 25 | 155 | 244 | 3 | 9 | 10w |
| 6 | 25 | 151 | 250 | 4 | 6 | 8w |
| 7 | 27 | 160 | 260 | 4 | 7 | 12w |
| Average | 24 | 153 | 223 | 4 | 8 | 8w 5d |
Table 2.
Patient subjective satisfaction score (PSSS) in patients post relaxation laryngoplasty
| S no | Age | Pre-op fundamental frequency | Post-op fundamental frequency | PSSS (Pre) | PSSS (Post) | Recovery time |
|---|---|---|---|---|---|---|
| 1 | 31 | 200 | 118 | 5 | 8 | 3w |
| 2 | 30 | 176 | 120 | 4 | 9 | 4w |
| 3 | 35 | 285 | 110 | 4 | 8 | 3w |
| 4 | 27 | 173 | 93 | 3 | 10 | 4w |
| 5 | 35 | 253 | 121 | 5 | 9 | 4w |
| 6 | 33 | 212 | 156 | 4 | 10 | 3w |
| 7 | 33 | 214 | 80 | 2 | 9 | 3w |
| Average | 32 | 216 | 114 | 4 | 9 | 3w 4d |
Table 3.
Tabulation of mean with standard deviation (SD) of pitch pre-operative in comparison to post-operative with the p-value in patients who underwent Wendler Glottoplasty
| Pre op | Post op | P value * | |
|---|---|---|---|
| pitch increase | 153.85 ± 7.104 | 223 ± 29.77 | P = 0.00057 S |
| PSS | 3.71 ± 0.48 | 7.28 ± 1.704 | P = 0.0025 S |
*t test was used for comparison
Table 4.
Tabulation of mean with standard deviation (SD) of pitch pre-operative in comparison to post-operative with the p-value in patients who underwent RLFS
| Pre op | Post op | P value * | |
|---|---|---|---|
| Pitch decrease | 216.14 ± 40.52 | 114 ± 24.07 | P = 0.00405 S |
| PSS | 3.85 ± 1.069 | 9 ± 0.82 | P = 0.00048 S |
*t test was used for comparison
Discussion
The aspects of voice that aid in identifying it to a particular gender include fundamental frequency, range of frequency, loudness, quality, resonance, and intonation pattern. The fundamental frequency (F0) pitch of voice is an important correlate with gender identification and pitch depends on the mass, tension, length, stiffness of vocal folds, and subglottal pressure. ST as a modality for voice modification has the advantage of being non-invasive and aids in modifying vocal behaviors. However, occasionally patients are not satisfied with ST alone [2]. The goal of our study was to assess the effectiveness of pitch-raising surgery performed on MTF transpersons and of pitch lowering by RFLS in puberphonic patients.
Failure of the voice to break at puberty and the inability of the pubescent to acquire the newly developed adult vocal pitch results in a vocal disorder that is referred to as puberphonia or mutational dysphonia [3]. The change in vocal pitch during puberty is secondary to the descent of the larynx and the developmental growth of its musculoskeletal structures [4]. In puberphonic patients, the backward pressure on the thyroid cartilage during phonation counteracts the tension of the cricothyroid muscle thus lowering the pitch [4]. This is known as a positive Gutzmann test, which is used as a diagnostic test [5]. One of the reasons for poor results for laryngeal compliance with therapy may be attributed to the hardening of the cartilage in the fourth decade of life [6, 7]. Therefore, relatively older patients are often candidates for RLFS [6, 7]. Isshiki proposed that all patients should undergo voice therapy for 3 to 6 months as a trial regardless of their examination findings and stated that only those patients who are not permanently cured and request surgery should be taken up [8]. In our voice clinic, patients over 25 years are given therapy for a minimum of 3 months and younger patients are given therapy for 6 months to 1 year before surgical consideration. The average age of the operated patients of puberphonia in our study sample was 32 years and they had received therapy for over 3 months. The average drop in pitch at 1 year in these patients was 102 Hz from 216 (216.14 ± 40.52) to 114 (114 ± 24.07) with p = 0.004 and there were no early or late postoperative surgical complications with PSSS improving from 4 to 9 with p = 0.0004 at I year postoperatively. Slavit et al. [9] in their study, reported a decrease in pitch in two patients from 221, 151 to 170, and 137 respectively. In another study done on seven patients, the mean F0 was improved from 187 to 104 Hz (p < 0.001) [5]. According to a study done by Nakumara et al. of only three patients, preoperative speaking F0 in these patients were 174.6, 170.2, and 180.0 Hz, which dropped to 106.9, 115.4, and 87.5 Hz, respectively [6]. With RFLS the drop in pitch is immediate and an initial breathy voice is expected. After a week of voice rest, strengthening and stability are focused upon with ST. With combined efforts, good results can be attained in a short span of 3 to 4 weeks. Our seven patients took on an average of 3 weeks and 4 days to complete healing. The result is a stable and highly satisfactory low-pitched voice. The mean PSSS improved from 4 (3.85 ± 1.069) to a fairly impressive 9 (9 ± 0.82) in patients who underwent RFLS. The scoring system has considered patients' overall satisfaction quotient. Type 3 thyroplasty was first reported by Isshiki [10] in the 1970s who developed the functional concept of laryngeal framework surgery as this surgery effectively relaxes the tension on the vocal folds it has been called Relaxation Laryngoplasty by the European Laryngological Association classification system [8].
Several surgical methods for pitch increase have been described such as cricothyroid approximation and anterior commissure advancement [11–13]. One of the oldest and still fairly used surgical methods known in literature is the cricothyroid approximation (CTA) described by Isshiki et al. [14] which had various modifications [15]. It produced good results initially but they failed to sustain as the pitch decline was observed within 6–18 months reportedly due to the loosening of sutures. LeJeune et al. [16] introduced anterior commissure advancement in 1983, further modified by Tucker et al. [17]. This procedure has the disadvantage of an external approach and additional accentuation of the thyroid prominence. Orloff et al. described a laser-assisted voice adjustment (LAVA) in which the vocal folds were vaporized using a laser [18].
Currently, popularly used is the endoscopic glottoplasty technique, described by Wendler in 1990, which consists of deep epithelialization and full-thickness suturing of the anterior portion of the vocal folds, forming an anterior web. In Kim’s technique, the retro positioning of the anterior commissure is achieved by taking the anterior most suture inferiorly which causes a 50% reduction of the vibrating length [19]. Since suturing is performed for web formation, WG patients possibly take a longer healing time, which was 8 weeks and 5 days in our study. To ensure the web does not open, some studies use various additional modalities such as fibrin glue [20], botulinum toxin [21] intraoperatively, and post-operative 10–14 days of voice rest. Like the study done by Remacle et al. [22] we too advice 2 weeks of post-operative voice rest. This post-operative voice rest was followed up with weekly ST for at least 3 months where other aspects like vocal hygiene, breathing, resonance, vocal smoothing, speech rhythm, intonation, and fundamental frequency were addressed.
In our study which was performed on seven transpersons, there was a significant improvement of 70 Hz from 153 (153.85 ± 7.104) Hz to 223 (223 ± 29.77) Hz with a p-value of 0.005 after the surgery and voice therapy. In a study done by Casado et al., there was an increase in F0 (106 Hz on average) after WG and voice therapy [23]. A study of seven patients who underwent wendlers showed an average rise in the fundamental frequency of 78 Hz [12, 13]. In another study of 31 patients who underwent wendlers, the mean F0 increased from preoperative 136 Hz to postoperative 206 Hz [24]. Mean F0 increased from 150 to 194 in a study done by Marc Remacle in 2011 [22]. Gross reported a decrease in the range for the lower frequencies, and an increase in the habitual frequency (mean pre 116.8 Hz, mean post 201.0 Hz) using the same technique [25]. Once good web formation has taken place, which is confirmed by flexible laryngoscopy, ST is started where other aspects like vocal hygiene, breathing, resonance, vocal smoothing, speech rhythm, intonation, and fundamental frequency were targeted. To be perceived as females, male-to-female transpersons need to increase their speaking F0 from about 120 Hz to 150–155 Hz [26]. A meta-analysis study done by Song et al. revealed that the endoscopic shortening group showed maximum elevation with a difference of 78.98 Hz followed by CTA which had a 44.97 Hz increase [27].
Dacakis noted that the interaction between F0, F0 range, intonation, and resonance ultimately determines the speaker's perception as female [2]. The PSSS of this group of patients increased from 4 (3.71 ± 0.48) to satisfactory 8 (7.28 ± 1.704) with p = 0.002. This is a fairly good score to attain in this group of patients but is less than the group of RFLS in our study. This might be attributed to all the accessory factors, which play a huge role in attaining the desirable feminine voice, unlike RFLS patients where only pitch needs tackling.
Conclusion
Puberphonia responds well to ST, though 20% of our patient’s needed surgery despite good compliance with therapy. This percentage is on the higher side, as a large bulk of patients are specifically referred to our voice clinic for surgery, following failed conservative measures. The results of RFLS are extremely gratifying, yielding a high PSSS with a pitch drop of more than 100 Hz, and this improvement was also translated into real-life connected speech. WG along with post-operative ST yielded a significant increase in the pitch with a good PSSS. The average healing time taken following WG was double the time taken for RFLS.
Acknowledgements
None
Funding
None.
Declarations
Conflict of interest
None.
Ethical Approval
Enclosed [Institutional ethics committee approval obtained]—BH-EC-0137.
Human Participants or Animals
Nil.
Informed Consent
Not Applicable.
Footnotes
Publisher's Note
Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.
References
- 1.Schwarz K, Fontanari AMV, Schneider MA, Borba Soll BM, da Silva DC, Spritzer PM, Kazumi YamagutiDorfman ME, Kuhl G, Costa AB, Cielo CA, Villas Bôas AP, Lobato MIR. Laryngeal surgical treatment in transgender women: a systematic review and meta-analysis. Laryngoscope. 2017;127(11):2596–2603. doi: 10.1002/lary.26692. [DOI] [PubMed] [Google Scholar]
- 2.Dacakis G. The role of voice therapy in male-to-female transsexuals. Curr Opin Otolaryngol Head Neck Surg. 2002;10(3):173–177. doi: 10.1097/00020840-200206000-00003. [DOI] [PubMed] [Google Scholar]
- 3.Hamdan AL, Khalifee E, Ghanem A, Jaffal H. Injection laryngoplasty in patients with puberphonia. J Voice. 2019;33(4):564–566. doi: 10.1016/j.jvoice.2018.02.017. [DOI] [PubMed] [Google Scholar]
- 4.Liang FY, Huang XM, Chen L, Huang YZ, Zhang XY, Su JH, Wang YJ, Yang JS, Zheng YQ, Mei XS, Guan Z. Voice therapy effect on mutational falsetto patients: a vocal aerodynamic study. J Voice. 2017;31(1):114.e1–114.e5. doi: 10.1016/j.jvoice.2016.03.017. [DOI] [PubMed] [Google Scholar]
- 5.Brodnitz FS. The pressure test in mutational voice disturbances. Ann Otol Rhinol Laryngol. 1958;67(1):235–240. doi: 10.1177/000348945806700121. [DOI] [PubMed] [Google Scholar]
- 6.Nakamura K, Tsukahara K, Watanabe Y, Komazawa D, Suzuki M. Type 3 thyroplasty for patients with mutational dysphonia. J Voice. 2013;27(5):650–654. doi: 10.1016/j.jvoice.2013.05.001. [DOI] [PubMed] [Google Scholar]
- 7.Dagli M, Sati I, Acar A, et al. Mutational falsetto: intervention outcomes in 45 patients. J Laryngol Otol. 2008;122:277–281. doi: 10.1017/S0022215107008791. [DOI] [PubMed] [Google Scholar]
- 8.Friedrich G, de Jong FI, Mahieu HF, Benninger MS, Isshiki N. Laryngeal framework surgery: a proposal for classification and nomenclature by the phonosurgery committee of the european laryngological society. Eur Arch Otorhinolaryngol. 2001;258(8):389–396. doi: 10.1007/s004050100375. [DOI] [PubMed] [Google Scholar]
- 9.Slavit DH, Maragos NE. Arytenoid adduction and type I thyroplasty in the treatment of aphonia. J Voice. 1994;8:84–91. doi: 10.1016/S0892-1997(05)80324-8. [DOI] [PubMed] [Google Scholar]
- 10.Isshiki N, Morita H, Okamura H, Hiramoto M. Thyroplasty as a new phonosurgical technique. Acta Otolaryngol (Stockh) 1974;78:451–457. doi: 10.3109/00016487409126379. [DOI] [PubMed] [Google Scholar]
- 11.Pasternak K, Francis DO. An update on treatment of 518 voice-gender incongruences by otolaryngologists and speech- 519 language pathologists. Curr Opin Otolaryngol Head Neck Surg. 2019;520(27):475–481. doi: 10.1097/MOO.0000000000000582. [DOI] [PubMed] [Google Scholar]
- 12.Aires MM, de Vasconcelos D, Lucena JA, Gomes AOC, Moraes BT. Effect of wendler glottoplasty on voice and quality of life of transgender women. Braz J Otorhinolaryngol. 2021 doi: 10.1016/j.bjorl.2021.06.010. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 13.Isshiki N, Taira T, Tanabe M. Surgical alteration of the vocal pitch. J Otolaryngol. 1983;12:5. [PubMed] [Google Scholar]
- 14.Tucker HM. Anterior commissure laryngoplasty for adjustment of vocal fold tension. Ann Otol Rhinol Laryngol. 1985;94:547–549. doi: 10.1177/000348948509400604. [DOI] [PubMed] [Google Scholar]
- 15.Remacle M, Matar N, Verduyckt I, et al. Relaxation thyroplasty for mutational falsetto treatment. Ann Otol Rhinol Laryngol. 2010;119:105–109. doi: 10.1177/000348941011900207. [DOI] [PubMed] [Google Scholar]
- 16.Lejeune FE, Guice CE, Samuels PM. Early experiences with vocal ligament tightening. Ann Otol Rhinol Laryngol. 1983;92:475–477. doi: 10.1177/000348948309200513. [DOI] [PubMed] [Google Scholar]
- 17.Tucker HM. Anterior commissure laryngoplasty for adjustment of vocal fold tension. Ann Otol Rhinol Laryngol. 1988;97:547–549. doi: 10.1177/000348948509400604. [DOI] [PubMed] [Google Scholar]
- 18.Orloff LA, Mann AP, Damrose JF, Goldman SN. Laser-assisted voice adjustment (LAVA) in transsexuals. Laryngoscope. 2006;116:655–660. doi: 10.1097/01.mlg.0000205198.65797.59. [DOI] [PubMed] [Google Scholar]
- 19.Kim HT. Vocal feminization for transgender women: current strategies and patient perspectives. Int J Gen Med. 2020;12(13):43–52. doi: 10.2147/IJGM.S205102.PMID:32104050;PMCID:PMC7024865. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 20.Yılmaz T, Kuşçu O, Sözen T, Süslü AE. Anterior glottic web formation for voice feminization: experience of 27 patients. J Voice. 2017;31(6):757–762. doi: 10.1016/j.jvoice.2017.03.006. [DOI] [PubMed] [Google Scholar]
- 21.Geneid A, Rihkanen H, Kinnari TJ. Long-term outcome of endoscopic shortening and stiffening of the vocal folds to raise the pitch. Eur Arch Otorhinolaryngol. 2015;272:3751–3756. doi: 10.1007/s00405-015-3721-7. [DOI] [PubMed] [Google Scholar]
- 22.Remacle M, Matar N, Morsomme D, et al. Glottoplasty for male-to-female transsexualism: voice results. J Voice. 2011;25:120–123. doi: 10.1016/j.jvoice.2009.07.004. [DOI] [PubMed] [Google Scholar]
- 23.Casado JC, O’Connor C, Angulo MS, et al. Wendler glottoplasty and voice-therapy in male-to-female transsexuals: results in pre and post-surgery assessment. Acta Otorrinolaringol Esp. 2016;67:83–92. doi: 10.1016/j.otorri.2015.02.003. [DOI] [PubMed] [Google Scholar]
- 24.Mastronikolis NS, Remacle M, Biagini M, et al. Wendler glottoplasty: an effective pitch raising surgery in male-to-female transsexuals. J Voice. 2013;27:516–522. doi: 10.1016/j.jvoice.2013.04.004. [DOI] [PubMed] [Google Scholar]
- 25.Gross M. Pitch-raising surgery in male-to-female transsexuals. J Voice. 1999;13:246–250. doi: 10.1016/S0892-1997(99)80028-9. [DOI] [PubMed] [Google Scholar]
- 26.Van Damme S, Cosyns M, Deman S, Van den Eede Z, Van Borsel J. The effectiveness of pitch-raising surgery in male-to-female transsexuals: a systematic review. J Voice. 2017;31(2):244.e1–244.e5. doi: 10.1016/j.jvoice.2016.04.002. [DOI] [PubMed] [Google Scholar]
- 27.Song TE, Jiang N. Transgender phonosurgery: a systematic review and meta-analysis. Otolaryngol Head Neck Surg. 2017;156(5):803–808. doi: 10.1177/0194599817697050. [DOI] [PubMed] [Google Scholar]