Muscle type of palatopharyngeal muscle in children with severe obstructive sleep apnea

Abstract

Study Objectives:

To investigate the fiber-type distribution in palatopharyngeal muscle via adenosine triphosphatase and quantitative real-time polymerase chain reaction in children with severe obstructive sleep apnea (OSA).

Methods:

Study participants were 12 children with severe OSA and 15 children with simple snoring as the control group. Both groups were diagnosed by polysomnography and treated with tonsillectomy. The samples of palatopharyngeus muscle were studied under adenosine triphosphatase staining and quantitative real-time polymerase chain reaction to classify the different fiber types.

Results:

There were no differences in baseline age, body mass index, tonsil size, or sleep stage constitution between the 2 groups. Dominance (>60%) of type I fiber was observed both in children with simple snoring (3/15, 20%) and in those with severe OSA (1/12, 8.3%) via adenosine triphosphatase staining. Predominance of type II fibers was seen in 3/15 (20%) in the control group and 6/12 (50%) in the severe OSA group, respectively. Type grouping was also seen in 8/15 (53.3%) in non-OSA and 6/12 (50%) in severe OSA groups, respectively. There was no difference in distribution of subtype fibers assessed by quantitative real-time polymerase chain reaction between the 2 groups; the mean percentages of type I fibers were 25.8% ± 19.5% and 20.9% ± 16.6%, respectively (P > .05), similar to type IIa fibers (35.2% ± 23.4% and 40.9% ± 28.8%) (P > .05). There was a decrease in the percentage of type I fibers between children younger and older than 12 years (P < 0.05), although this was not due to OSA (P > 0.05).

Conclusions:

There were no specific changes via adenosine triphosphatase staining or a difference in distribution of subtype fibers via quantitative real-time polymerase chain reaction between children with severe pediatric OSA and those with simple snoring, whereas the percentage of type I fiber decreased dynamically due to age but not OSA.

Clinical Trial Registration:

Registry: Chinese Clinical Trials Registry; Name: A study of the mechanism of the conversion of upper airway expasion muscle’s fiver types of OSA patient which may be mediated by estrogen-related receptor alpha; URL: https://www.chictr.org.cn/showproj.aspx?proj=6144; Identifier: ChiCTR-CCC-13003415.

Citation:

Chen H, Huang X, Ye Y, Luo Y, Huang Y, Li X. Muscle type of palatopharyngeal muscle in children with severe obstructive sleep apnea. J Clin Sleep Med. 2020;16(9):1523–1529.

BRIEF SUMMARY

Current Knowledge/Study Rationale: We investigated the muscle type of upper airway via adenosine triphosphatase staining and quantitative real-time polymerase chain reaction in severe childhood obstructive sleep apnea and simple snoring group.

Study Impact: There were no specific changes via adenosine triphosphatase staining, or a difference in distribution of subtype fibers via quantitative real-time polymerase chain reaction between the 2 groups, whereas the percentage of type I fibers decreased dynamically due to age but not obstructive sleep apnea.

INTRODUCTION

Obstructive sleep apnea (OSA) is a common breathing disorder characterized by repetitive narrowing and closure of the upper airway during sleep. One of the suggested mechanisms involved in OSA is the neuromuscular abnormality of the palatal muscles. The function of the upper airway muscle is of vital importance to maintain the stability and patency of the upper airway during inspiration, a muscle dysfunction that might be one of the contributors to OSA.

The function of mammalian skeletal muscle is determined by different muscle fiber types, which have variable actions with regard to the rate of force production, resistance to fatigue, and energy metabolism. Our previous study demonstrated that the rate of type I slow-twitch myofiber (Myh7) was decreased in the palatopharyngeal muscles of patients with OSA.¹ A reduced rate of type I fibers can lead to fatigue and greater susceptibility to pharyngeal collapse during sleep.

It was first proposed in 1991 that individuals with OSA and/or severe snoring had more type IIA and less type IIB and type I fibers than nonapneic snorers in the medial pharyngeal constrictor muscle.² A substantial body of literature then explored neuromuscular changes in OSA, but research on the role of muscle fiber in the pathophysiology and pathogenesis of childhood OSA is thus far limited.

Therefore, the present study was designed to examine if the same fiber-type transition occurred in childhood OSA. At the same time, even to the experienced pathologist, it is still difficult to identify all subtypes of pharyngeal muscles.³ We therefore investigated the muscle fiber types of the upper airway via adenosine triphosphatase (ATPase) staining and quantitative real-time polymerase chain reaction (qRT-PCR), 2 techniques that could confirm the subtypes of muscle fibers more accurately, in children with severe childhood OSA and those with simple snoring. Furthermore, we believe that this is the first study to describe the changes in muscle types from childhood to adolescence. A clearer understanding of how muscle-type transition influences the pathophysiology of childhood OSA could enable better interventions for children with postoperative residual elevation in their apnea-hypopnea index (AHI).

METHODS

Participants

Individuals with moderate to severe OSA (n = 12) and non-OSA simple snoring (n = 15) aged 3–16 years were enrolled from September 2014 to July 2016. Patients were evaluated in a sleep laboratory for the complaint of snoring and/or sleepiness and with subsequent evaluation of sleep apnea after full polysomnography. None of the patients had clinical complaints or physical findings of neuropathy or neuromuscular disorder. Our study was carried out in accordance with the guidelines of the Ethics Committee of Nanfang Hospital and approved by the institutional review board of the instruction. The study conforms to the Declaration of Helsinki. The project was approved by Chinese Clinical Trials Registry (ChiCTR-CCC-13003415) on 31 May 2013.

Sleep study

All patients underwent initial diagnostic polysomnography using a computerized system (Embla, Middleton, WI). Continuous recordings were taken with an international 10–20 electrode placement system, including eye movements, chin electromyogram, and electrocardiogram. Sleep and respiratory events were scored manually according to standard criteria. Respiratory efforts were monitored with abdominal and thoracic bands. An additional signal of respiratory effort (ie, pulse transit time) was recorded concurrently. Oxygen saturation was measured using a digital pulse oximeter. AHI was calculated and defined as the number of apneas and hypopneas per hour of sleep (full polysomnography).

Sleep parameters included the AHI, duration of desaturation, obstructive apnea index (OAI), and lowest and mean oxygen saturation (SpO₂). We defined abnormality as AHI >5 events/h or OAI >1 events/h, and hypoxia as the lowest oxygen saturation <92%. The diagnostic criteria for childhood OSA were based on the 2007 Chinese guidelines for the diagnosis and treatment of childhood OSA.⁴ Sleep hypopnea was defined as airflow being reduced to more than 50% of baseline in association with a 3% decreased oxygen saturation or arousal. OSA severity was quantified as follows: AHI <5 and OAI <1 events/h (non-OSA), 5<AHI<10 or 1<OAI<5 events/h (mild), 10<AHI<20 or 5<OAI<10 events/h (moderate), and AHI ≥20 or OAI >10 events/h (severe).

Enzyme-histochemical staining

The muscle biopsy samples were obtained during tonsillectomy under general anesthesia without local anesthetic. All surgeries were done by residents. Each specimen was taken from the palatopharyngeus, quickly frozen in liquid nitrogen, and stored in a deep freezer at −750°C. Serial cross-sections, 10-μm thick, were cut at −200°C in a cryostat microtome and mounted on glass slides.

The slides were stained with hematoxylin–eosin and myofibrillar ATPase after alkaline (pH 9.6) and acid (pH 4.6 and PH 4.3) preincubation. According to the ATPase staining, we primarily classified type I from type II muscle fiber types.

We observed fiber type I or II predominance (>60%), type grouping (≥12 fibers of the same type), grouped or fascicular atrophy, and angulated fibers. Four randomly chosen areas in each section of every muscle sample stained with hematoxylin–eosin and ATPase at pH 9.6, PH 4.6, and PH 4.3 were scanned with a video camera.

RNA extraction and qRT-PCR

Total RNA was extracted from the samples using RNAiso Plus (Takara, Shiga, Japan), according to the manufacturer’s instructions. RNA concentration was measured by spectrophotometry. After cDNA synthesis (All-in-One First-Strand cDNA Synthesis kit; GeneCopoeia, Inc, USA), qRT-PCR was performed using the All-in-One qPCR Mix (GeneCopoeia, Inc, Rockville, MD) on the ABI 7500HT System (Applied Biosystems). Conditions for qRT-PCR were as follows: 95°C for 10 minutes followed by 40 cycles of 95°C for 10 seconds, 60°C for 20 seconds, and 72°C for 34 seconds. The specificity of each qRT-PCR reaction was verified by melting curve analysis. Duplicate reactions were run for each sample. All samples were normalized based on the expression levels of 18S ribosomal RNA and fold-changes were calculated using the 2^ΔΔCT method. The percentage of MyhX in the sample was calculated as follows (where X = 1, 2, 4, or 7): [fold(MyhX)]/[fold(Myh7) + fold(Myh2) + fold(Myh1) + fold(Myh4)] × 100.

RESULTS

Clinical data for the 2 groups

There was no difference in baseline age, body mass index, tonsil size, apnea index, or sleep structure between children with severe OSA and non-OSA control children (see Table 1); however, there were differences in Mallampati scores, AHI, SpO₂ nadir, and average SpO₂ (%) between the 2 groups.

Table 1.

Clinical characteristics and sleep respiratory findings in PSG in children with severe OSA and non-OSA control children.

								Sleep Stage, %
Group and Age, years	Tonsil Size	Mallampati Score	BMI, kg/m2	AHI, events/h	AI, events/h	Nadir SpO2, %	Average SpO2, %	Stage I	Stage II	Stage III	REM
Non-OSA snoring
6.5	3	1	15.6	1.1	0.7	90	97.9	0.1	36.3	44.4	19.2
3.5	2	1	18.5	2.4	1.6	89	98.5	19.9	11.9	55.6	12.7
6	3	1	15.5	4	0.1	92	94.9	0	0	91.4	8.6
8	2	2	31.2	4.2	0.8	93	95	2.2	40.9	38.6	18.2
6	2	1	13.9	4.3	0.1	90	96.7	1.6	8	74	12.9
7	4	1	13	2.8	0.6	94	98	1.9	16.6	61.6	19.9
5	3	1	17.5	2.8	0.9	95	97.4	0.3	28.6	40.6	30.5
3	3	1	15.4	3	0.3	94	97.5	0.6	6.7	68.5	24.7
4	4	1	14.7	4.7	0.7	93	97.1	0.9	17	57.7	24.5
5.8	3	1	12.9	4.9	0.6	91	96.9	2.6	25.4	37.6	34.3
15	3	2	17.8	0.2	0.1	93	97.5	9.7	23.5	46.7	20.1
13	3	1	16.8	1.4	0.8	91	95	2.3	16.6	66.8	14.3
16	4	1	19.1	3	0	92	95.8	3.7	66.4	22.9	6.9
12	2	1	14.9	2.9	0.2	91	96.3	8.2	19.9	43.5	28.4
12	3	1	25.7	3.6	0.2	92	96.7	14.9	24.1	37.2	23.9
Severe OSA
4.7	3	1	17.7	11.2	2.5	84	96.1	0.2	27.2	62.5	10
5	3	2	16.5	13.3	7.7	79	95.5	0.8	26.9	46.1	26.1
9	4	1	20.7	15.3	4	73	95.3	0.8	10.5	71.2	17.5
6	2	2	17.7	17	9	71	96.8	9.8	20.5	34.2	35.5
5	2	1	14.1	19.4	17.2	67	95.8	0.3	19.1	57	23.6
6	3	1	15.9	23	10.2	62	96.1	6.7	18.9	45.9	28.5
4	2	2	16.9	25.1	0.7	71	94.1	0.9	34.7	47.8	16.6
5	3	1	14.2	36.4	15.3	66	95.6	0.8	28.4	47.9	22.9
12	2	3	30.4	81.8	73.4	51	84.5	4.5	68.8	16.7	10.1
12	3	2	34.5	10.2	0.3	82	96	2.8	43.9	29.7	23.5
12	2	3	20.8	11.4	8.3	60	93.2	6.9	49	18.6	25.5
13	2	1	26.8	24.5	4.3	86	96.0	3.4	27.3	40.2	29
P = .807	P = .201	P = .004	P = .186	P = .004	P = .056	P = .000	P = .026	P = .461	P = .187	P = .174	P = .424

BMI = body mass index, AHI = apnea-hypopnea index, AI = apnea index, OSA = obstructive sleep apnea, PSG = polysomnography, REM = rapid eye movement, SpO₂ = oxygen saturation.

Fiber-type distribution in enzyme-histochemical staining

Differentiation of the 4 subtypes of fibers via ATPase staining was difficult. Dominance (>60%) of type I fibers was observed both in the matched control group (3/15, 20%; Figure 1A) and in the group with severe OSA (1/12, 8.3%) via ATPase 4.3 staining. There was also a predominance of type II fibers in both groups: 3/15 (20%; Figure 1B) in the non-OSA group and 6/12 (50%; Figure 2A) in the group with severe OSA. A normal distribution of different fiber types was more common in the non-OSA snoring group (Figure 2B). Clusters of fibers of the same fiber type (type grouping) were also observed in both groups: 5/15 (30.3%; Figure 1B) in the non-OSA group and 5/12 (41.6%; Figure 2A) in the group with severe OSA.

Figure 1. Cross-sections from a palatopharyngeus muscle obtained from a patient without OSA.

(A) Predominance of type I, ATPase PH 4.3 staining, original magnification ×100. (B) Predominance of type II and clusters of fibers of the same type (type grouping) were observed from a patient without OSA with ATPase pH 4.6, original magnification ×100. OSA = obstructive sleep apnea.

Figure 2. Cross-sections from palatopharyngeus muscle obtained from severe OSA (A) and non OSA snoring (B).

Sections are stained for ATPase at pH 4.3, original magnification ×100. (A) Type II fibers in dominant and clusters of fibers of the same type (type grouping) were observed in severe OSA group. (B) Normal distribution of different fiber types in non-OSA group.

Fiber-type distribution via qRT-PCR

There were no significant differences in the distribution of subtype fibers assessed by qRT-PCR between the control group and the group with severe OSA (Table 2). The mean percentages of type I fibers were 25.8% ± 19.5% and 20.9% ± 16.6%, respectively (P > 0.05), similar to those of type IIa fibers (35.2% ± 23.4% and 40.9% ± 28.8%; P > .05), type IIb fibers (14.8% ± 21.2% and 13.0% ± 19.1%; P > 0.05), and type IIx fibers (24.3% ± 11.5% and 25.1% ± 15.5%; P > 0.05). Not unexpectedly, the same trend appeared for the percentage of oxidative fibers (sum of type I and IIa) and glycolytic fibers (sum of type IIb and IIx).

Table 2.

qRT-PCR findings for palatopharyngeus muscle types in non-OSA control children and children with severe OSA.

		Muscle Fiber Type, %
Group and Age, years	AHI, events/h	I	IIa	IIb	IIx	Oxidative, %	Glycolytic, %
Non-OSA snoring
6.5	1.1	29.83	42.78	18.84	8.54	72.61	27.38
3.5	2.4	1.92	2.49	56.92	38.67	4.41	95.59
6	4	20.16	52.11	1.14	26.59	72.27	27.73
8	4.2	21.24	18.25	36.72	23.79	39.49	60.51
6	4.3	18.42	32.16	7.84	41.58	50.58	49.42
7	2.8	49.2	36.88	0.61	13.31	86.08	13.92
5	2.8	21.71	48.62	1.54	28.13	70.33	29.67
3	3	34.42	15.47	9.08	41.03	49.89	50.11
4	4.7	76.62	4.06	6.53	12.79	80.68	19.32
5.8	4.9	29.78	45.25	3.72	21.25	75.03	24.97
15	0.2	12.97	58.75	3.1	25.19	71.72	28.29
13	1.4	2.69	1.28	66.07	29.96	3.97	96.03
16	3	42.84	38.14	3.47	15.54	80.98	19.01
12	2.9	8.77	85.14	0.86	5.23	93.91	6.09
12	3.6	16.21	45.88	4.96	32.95	62.09	37.91
8.2 ± 4.3*	3.0 ± 1.4	25.8 ± 19.5	35.2 ± 23.4	14.8 ± 21.2	24.3 ± 11.5	60.9 ± 27.1	39.1 ± 27.1
Severe OSA
4.7	11.2	5.23	2.88	48.78	43.1	8.11	91.88
5	13.3	10	65.06	2.17	22.77	75.06	24.94
9	15.3	30.36	54.26	1.74	13.64	84.62	15.38
6	17	26.29	33.76	1.31	38.65	60.05	39.96
5	19.4	22.13	63.01	2.61	12.25	85.14	14.86
6	23	34.63	15.01	0.92	49.43	49.64	50.35
4	25.1	38.95	3.99	26.35	30.71	42.94	57.06
5	36.4	55.54	37.28	0.08	7.09	92.82	7.17
12	81.8	13.02	58.17	1.6	27.2	71.19	28.8
12	10.2	6.53	79.56	2.05	11.86	86.09	13.91
12	11.4	4.82	74.69	16.76	3.74	79.51	20.5
13	24.5	3.66	3.6	51.74	41	7.26	92.74
7.8 ± 3.5	24.1 ± 19.7	20.9 ± 16.6	40.9 ± 28.8	13.0 ± 19.1	25.1 ± 15.5	61.9 ± 29.5	38.1 ± 29.5
P = .807	P = .004	P = .498	P = .569	P = .826	P = .877	P = .933	P = .932

*Mean ± SD (all such values). Significant values appear in bold. AHI = apnea-hypopnea index, OSA = obstructive sleep apnea, qRT-PCR = quantitative real-time polymerase chain reaction.

Although our study did not identify a difference in fiber types related to the severity of OSA, when we explored the data further we found that the difference in type I muscle fibers decreases with age in both groups, although slightly, but not significantly, more in the OSA group. With 2-factor analysis of variance, there was a statistical difference in type I fibers between children younger and older than 12 years (P = .020 < .05), whereas there was no difference between the OSA and control groups (P = .458) (Table 3). Therefore, age but not OSA led to the decrease in type I fibers.

Table 3.

Fiber-type distribution in different age subgroups (2-factor ANOVA).

	Muscle Fiber Type, %
	I	IIa	IIb	IIx	Oxidative, %	Glycolytic, %
Control Age, years
3 to <12	30.3 ± 20.3	29.8 ± 18.4	14.3 ± 18.6	25.6 ± 12.0	60.1 ± 24.6	39.9 ± 24.6
12 to <16	16.7 ± 15.5	45.8 ± 30.6	15.7 ± 28.2	21.8 ± 11.4	62.5 ± 34.8	37.5 ± 24.6
OSA Age, years
3 to <12	27.9 ± 16.1	34.4 ± 25.2	10.5 ± 17.8	27.2 ± 15.7	62.3 ± 28.3	37.7 ± 28.3
12 to <16	7.0 ± 4.2	54.0 ± 34.8	18.0 ± 23.5	20.9 ± 16.5	61.0 ± 36.3	38.9 ± 36.3
P value
Age	.020	.096	.628	.348	.949	.949
OSA	.458	.555	.828	.877	.934	.934

Significant values appear in bold. ANOVA = analysis of variance, OSA = obstructive sleep apnea.

DISCUSSION

Traditional upper airway reconstructive procedures for OSA exclusively target the anatomy and structural volume ability of the upper airway; however, increasing evidence demonstrates that neuromuscular compensation during sleep plays a critical role in the sleep-disordered breathing pathophysiology of many patients with OSA.⁵ Residual OSA occurred in 38–55% of children after adenoid-tonsillectomy,^6,7 and surgical results of lingual tonsillectomy for pediatric OSA are also not uniformly successful.⁸ It appears that, beyond anatomic abnormalities, there are other contributors to childhood OSA that have not been well characterized and do not respond to surgery.⁹

It is not clear if upper airway myopathy is the primary cause of OSA or is secondary to pathophysiologic changes related to the presence of OSA.^10,11 However, it is widely accepted that remodeling of the upper airway muscles does occur in adult OSA.¹²

Previous histological studies in adults with OSA have shown chronic neurogenic changes, such as muscle fiber atrophy, in the upper airway dilator muscle.¹³ Upper airway muscles were hypothesized to be highly trained by chronic overnight loading and/or hypoxia. Sériès et al¹⁴ found that the protein content of musculus uvulae, total number of muscle fibers, the number and size of type IIA fibers, and total muscle fiber cross-sectional area were significantly greater in patients with OSA than in snorers. Another study by Sériès et al¹⁵ revealed that musculus uvulae had a greater proportion of type IIA fibers in patients with OSA than in snorers. This study also identified an increased proportion of type IIA muscle fibers in adult OSA,^2,16 which also included changes in capillary density, mitochondria content,¹⁷ and fiber-type grouping¹⁸; and increased proportion of small-sized fibers and frequency of fibers containing developmental MyHC isoforms,³ etc. However, whether the same muscle-type transition occurs in prepuberty or puberty-stage OSA remains unknown.

The study by Vuono et al¹⁹ demonstrated that histological findings in the palatopharyngeus, such as fiber-size variability, fiber-type grouping, and predominance of type II fibers, would not only be seen in individuals with primary snoring or apnea but also in children without sleeping disorders, and may represent normal rather than neurogenic or myopathic pathology. Subsequently, an animal model verified that early-life exposure to chronic intermittent hypoxia primed increased susceptibility to hypoxia-induced weakness in the sternohyoid muscle of adult rats.²⁰ We are not aware of another study that focuses on the childhood OSA muscle-type transition. Here we first report that, even in severe pediatric OSA, there were no differences in muscle-type distribution compared with a group with simple snoring.

The Penn State Child Cohort was a longitudinal study that confirmed that prepubertal OSA tended to resolve naturally during the transition to adolescence, and that primary snoring and mild sleep-disordered breathing did not appear to be strongly associated with progression to more severe sleep-disordered breathing.²¹ Due to the high frequency of spontaneous resolution of mild OSA,²² use of adenotonsillectomy was sharply decreased.²³

Despite this, concern remains that primary snoring will develop to OSA over time. Even with high-quality clinical data that do not demonstrate a risk of this transition,²⁴ we still lack histological proof. Our study found that there were no specific histological features differentiating these 2 groups, or a statistical difference in baseline and fiber-type distribution. Our study provides histological proof that there are no direct links between simple snoring and severe pediatric OSA. This may lead to a decrease in the use of surgery for pediatric simple snoring. As shown in Table 2, the distribution of fiber type varied sharply both in the control group and in the group with severe childhood OSA. Skeletal muscle fiber–type distribution is quite heterogeneous since the proportion of fiber types is primarily genetically determined.²⁵ Our data support a trend of transition of muscle-fiber type during the different stages of childhood. Of interest, a 10-year follow-up study regarded childhood and adolescent OSA as 2 distinct entities, with the latter more likely to persist into adulthood. A proportion of children with OSA, particularly girls, had complete resolution during transition to late adolescence or early adulthood.²²

Are the age-related changes we have identified due to repetitive intermittent hypopneas, which lead to a decrement in pharyngeal dilator function during sleep? Or are changes simply due to normal physiologic maturation with increasing age? These questions require further study. Our study provides initial information to direct investigations toward longitudinal studies. There was a decrease in the percentage of type I fiber between children younger and older than 12 years, although this was not due to OSA severity. Because there is no international agreement on diagnostic criteria for childhood OSA, a further understanding of dynamic muscle-type transition from prepubertal to adolescence is of vital important. The number of participants in our study was still limited, so our findings need further confirmation. In addition, the children with severe OSA in our study had a fairly modest mean AHI (24.1 ± 19.7 events/h), which may explain our negative results. However, we suggest that muscle fiber–type transition and characteristics of different age groups may be important considerations in the treatment decisions for pediatric OSA as our understanding grows.

DISCLOSURE STATEMENT

Conception and design of the paper: Drs. Huaihong Chen and Xiangpin Li. Analysis and interpretation: Yanqing Ye and Yunfang Luo. Drafting the manuscript for important intellectual content: Xiaoxing Huang and Yuanshou Huang. All authors have seen and approved the manuscript. This study was funded by the Guangdong Natural Science Foundation of China (grant no. 2018A030313805 to H. Chen) and the Clinical Research Start Up Program of Southern Medical University by High-level University Construction Funding of the Guangdong Provincial Department of Education (grant no. LC2016ZD011 to X. Li). The authors report no conflicts of interest.

ABBREVIATIONS

AHI

apnea-hypopnea index

ATPase

adenosine triphosphatase

OAI

obstructive apnea index

OSA

obstructive sleep apnea

qRT-PCR

quantitative real-time polymerase chain reaction

SpO₂

oxygen saturation