Abstract
Background:
Hypothyroidism has been associated with increased pulmonary morbidity and overall mortality. A systematic review was conducted to identify the prevalence and underlying mechanisms of respiratory problems among patients with thyroid insufficiency.
Methods:
PubMed and EMBASE databases were searched for relevant literature from January 1950 through January 2015 with the following study eligibility criteria: English-language publications; adult subclinical or overt hypothyroid patients; intervention, observational, or retrospective studies; and respiratory manifestations. The Preferred Reporting Items for Systematic reviews and Meta-Analyses statement was followed, and Cochrane's risk of bias tool was used.
Results:
A total of 1699 papers were screened by two independent authors for relevant titles. Of 109 relevant abstracts, 28 papers underwent full-text analyses, of which 22 were included in the review. Possible mechanisms explaining respiratory problems at multiple physiological levels were identified, such as the ventilator control system, diaphragmatic muscle function, pulmonary gas exchange, goiter caused upper airway obstruction, decreased capacity for energy transduction, and reduced glycolytic activity. Obstructive sleep apnea syndrome was found among 30% of newly diagnosed patients with overt hypothyroidism, and demonstrated reversibility following treatment. The evidence for or against a direct effect on pulmonary function was ambiguous. However, each of the above-mentioned areas was only dealt with in a limited number of studies. Therefore, it is not possible to draw any strong conclusions on any of these themes. Moreover, most studies were hampered by considerable risk of bias due for example to small numbers of patients, lack of control groups, randomization and blinding, and differences in body mass index, sex, and age between subjects and controls.
Conclusion:
Mechanistic data linking hypothyroidism and respiratory function are at best limited. This area of research is therefore open for retesting hypotheses, using appropriate study designs and methods.
Introduction
O
Based on the observed increased mortality and pulmonary comorbidity, this study aimed to investigate by a review of the literature the impact of hypothyroidism on pulmonary function and the respiratory system—an issue very sparsely addressed previously.
Methods
A systematic review was performed according to the Preferred Reporting Items for Systematic reviews and Meta-Analyses (PRISMA) statement (14). Methods as well as inclusion and exclusion criteria were specified in advance in a review protocol (access available at
English-language publications on the relationship between respiratory function and hypothyroidism, accessible in PubMed and EMBASE databases (January 1, 1950–January 31, 2015) were identified. The Population, Intervention, Comparison, Outcome and Study (PICOS) approach was used to set the research question (15). All studies dealing with overt or subclinical hypothyroid patients were considered if the outcome was related to the respiratory function. Both observational and interventional studies were accepted, irrespective of whether the design was retrospective, cross-sectional, or prospective. No comparator group was necessary. The last search was performed on February 20, 2015. The full search strategy is provided in the Appendix.
Literature search
Two authors (J.S. and K.W.) independently screened the titles and abstracts. They decided, by consensus, which articles to evaluate in full text. A search was made for outcomes related to neurological ventilatory control, pulmonary gas exchange, pulmonary function, respiratory strength, and sleep apnea. Only English language articles were included. The exclusion criteria were: case reports, expert opinions, letters, reviews, and studies conducted in pregnant patients or those with thyroid cancer. Multiple reports of the same set of data were assessed, and only the most representative or updated report was included. The reference lists of included full-text articles were screened for missing publications. The same two authors extracted data independently, according to a pre-specified data-collection sheet. Disagreements were resolved by discussion. The following data were extracted: type and design of the study, first authorship, country of origin, year of publication, number of patients, demographics, severity of hypothyroidism, type of intervention if relevant, time of post-intervention follow-up, and outcome parameters. The Cochrane Risk of Bias tool was used to assess the risk of bias in each of the included articles (16).
Results
A total of 1690 relevant titles were identified from PubMed and EMBASE, and nine from the reference lists of included papers (Fig. 1). Of these, 1672 papers were excluded after a review of titles and abstracts, leaving 28 papers for full-text evaluation. Six full-text papers failed to meet the inclusion criteria, resulting in 22 papers for the final analyses. No randomized double-blind placebo-controlled studies were identified.

Preferred Reporting Items for Systematic reviews and Meta-Analyses (PRISMA) flow chart of manuscript decision process.
Applying the Cochrane Risk of Bias tool across studies, a considerable risk of bias was found, as many studies failed to describe inclusion procedures, include control groups, randomize for treatment, blind observers or participants, or report data previously described in the method section (Table 1). As a consequence, no attempt was made to perform a meta-analysis.
?, unclear risk of bias; +, low risk of bias; −, high risk of bias.
Most of the included studies (Table 2) were small, with ≤50 individuals in 13/22 studies. Sixteen studies included overt hypothyroid patients only, and four studies included subclinical hypothyroid patients only. The two remaining studies included both groups of hypothyroid patients. There were 13 intervention studies, eight prospective observational studies, and one retrospective observational study.
Age: mean (years). TSH is given as the mean value (mIU/:) at time of diagnosis.
OB, observational; IN, interventional; RE, retrospective; HT, overt hypothyroidism; sHT, subclinical hypothyroidism; TSH, thyrotropin; LT4, levothyroxine; LT3, L-triiodothyronine; Quest, questionnaire; CO2, carbon dioxide measurements; CO, carbon monoxide; MVM, manovacumeter; PFT, pulmonary function test; FVC, forced vital capacity; FEV1, forced expiratory volume in 1 second; PSG, polysomnography; OSAS, obstructive sleep apnea syndrome; AHI, Apnea Hypopnea Index; RDI, Respiratory Disturbance Index; NA, not available.
Respiratory symptoms and ventilation
Six studies were identified addressing the effect of overt or subclinical hypothyroidism on respiratory symptoms and ventilation. Four of these had substantial risk of bias in their study design (Table 1), defined as not using randomization or allocation concealment or blinding. Among 124 and 20 patients with overt hypothyroidism, two observational studies with severe risk of bias (Table 1) found increased respiratory symptoms, including shortness of breath, sputum production, cough, wheezing, and airway hyper-responsiveness, with an odds ratio of 2.7–3.5, when compared with 1346 healthy controls (17,18).
It has been debated whether pleural effusion, due to overt hypothyroidism, causes some of the respiratory symptoms in patients with thyroid failure. A single study with high risk of bias (Table 1) found that the effusions in such patients were primarily due to diseases other than hypothyroidism (19).
A study of overt hypothyroid patients found that the response to hypercapnia and hypoxia was reduced initially, improved seven days after either L-triiodothyronine (LT3) or levothyroxine (LT4) treatment, and normalized in most patients after 12–24 weeks of LT4 treatment only (20). That study, however, was hampered by 25% of the patients being lost to follow-up. Another study failed to reproduce these results (21), which may be due to a sample size that was half of that used in the previous study. In a non-blinded study, including patients with overt or subclinical hypothyroidism, the end-tidal CO2 was reduced (22). This finding is surprising, but most likely explained by hyperventilation at the time of examination.
Based on these data, no clear conclusions can be made regarding the impact of hypothyroidism on respiratory symptoms and ventilation. The included studies all use different assessment techniques, and are hampered by substantial risk of bias. It is therefore not possible to draw any conclusions on this topic. Only two studies have addressed the effects of LT4 substitution on ventilation (20,21). Both studies suffer from a high risk of observer bias, and they present opposing results, which renders a conclusion impossible (Table 3).
Moderate level of evidence: further research is likely to have an important impact on our confidence in the estimate of effect and may change the estimate. Low level of evidence: further research is very likely to have an important impact on our confidence in the estimate of effect and is likely to change the estimate.
Pulmonary function
The results from the 11 studies addressing the influence of overt or subclinical hypothyroidism on pulmonary function are contradictory. Of the 11 studies, nine have study designs with considerable risk of bias (Table 1). Three studies addressed the impact of overt hypothyroidism on the diaphragmatic and abdominal muscle strength, and offer contradictory results (21,23,24). Two studies—one with 43 patients (23) and another with 24 patients (24)—found that the diaphragmatic muscle strength improved with either LT4 or LT3 treatment. A third study, including 20 patients, found no improvement after LT4 treatment (21). All three studies included a limited number of participants and can therefore not be the foundation for any strong conclusions (21,23,24).
In subclinical hypothyroidism, a single study found a reduced diaphragmatic inspiratory and expiratory strength, but the results were invalidated by 3.7 times more men being included in the control group than in the study group (25). As for the diaphragmatic inspiratory strength, no change in this parameter has been demonstrated in this group of patients (26).
Two studies comprising 267 and 120 patients suffering from subclinical hypothyroidism found the forced vital capacity (FVC) reduced by >250 mL and the forced expiratory volume in 1 second (FEV1) reduced by 190 mL compared with healthy participants (25,27). A study of 20 overt hypothyroid patients showed similar findings compared with healthy participants (28). Although the results may seem unequivocal, they should be interpreted with caution, as the study and control groups were not comparable with respect to sex, age, and body mass index (BMI).
Apart from these three reports, the majority of studies showed no impact of thyroid dysfunction on pulmonary function when using pulmonary function tests (18,24,29 –31), most likely because of small sample sizes (21–45 patients). With a few exceptions (18,24,30), the lack of a control group is clearly another limitation of these studies. One study with 43 overt hypothyroid patients found that the pulmonary function improved after initiation of LT4, despite the fact that it was not affected at baseline (23). Other studies with fewer patients could not reproduce this finding (24,29 –31), which might be due to limited power.
As the majority of the studies have serious methodological limitations, it is difficult, if not impossible, to draw any clear conclusions regarding the impact of hypothyroidism or the effect of LT4 therapy on the diaphragmatic muscle strength and pulmonary function (Table 3).
Obstructive sleep apnea syndrome
Seven studies were identified that investigated the impact of overt or subclinical hypothyroidism on obstructive sleep apnea syndrome (OSAS). Three of the seven studies carried a significant risk of bias. Comparison across studies is difficult, as study designs, populations, and the employed techniques varied.
Nocturnal breathing abnormalities, such as restless sleep, snoring, choking, and in severe cases apnea periods, occurred among 25–50% of patients with overt hypothyroidism (32 –35). One study found that 30% of patients with recently diagnosed primary overt hypothyroidism suffered from OSAS, according to well-defined criteria (34). In these patients, OSAS was reversed by LT4 treatment (34). The other studies were limited by differences in BMI between control and patient groups, presence of goiter, small patient numbers, or missing data on thyroid hormone levels. After LT4 replacement therapy, all interventional studies demonstrated a significant reduction in apnea periods, oxygen desaturation events, snoring, and choking (33 –37). In subclinical hypothyroidism, one study from a specialized sleep clinic found that 53% of the patients had OSAS (38). However, a similar high rate of OSAS was found among euthyroid subjects, probably reflecting selection bias in that study.
It can be concluded that overt hypothyroidism seems to be linked to OSAS, with improvement after LT4 substitution, as shown in four of the six studies (Table 3). The evidence for subclinical hypothyroidism being linked with OSAS is vaguer, as only one study with serious methodological limitations addressed this issue.
Discussion
A systematic review was undertaken with the aim of elucidating the type and magnitude of respiratory problems among patients with hypothyroidism, and, if possible, of identifying the underlying physiological mechanisms. Unfortunately, many of the studies on which this review was based are hampered by considerable heterogeneity, generally low participant numbers, and high risk of observer and selection bias. Most credence was given to the studies with the least bias, and these were weighted highest in the conclusions, which, however, are still rather vague.
Overt and subclinical hypothyroidism
Overt hypothyroidism may be associated with an increased risk of respiratory symptoms (19). Such patients may also have a decreased control of breathing in response to hypercapnia and hypoxia (20), diminished diaphragmatic muscle strength (23,24), and higher propensity to develop sleep apnea (34). These statements are, however, associated with great uncertainty, as the evidence is sparse. Additionally, the few studies conducted within this area are hampered by considerable observer and selection bias, as they rarely use blinding or randomization in their designs.
In case of subclinical hypothyroidism, patients may have reduced diaphragmatic muscle strength and lower FVC and FEV1, according to two studies (25,27), but these results should be interpreted with caution, since the male/female ratio was higher in the control groups. Whether patients with subclinical hypothyroidism suffer from unrecognized OSAS is unclarified (38).
Thyroid hormone substitution
In overt hypothyroid patients, both LT3 and LT4, taken separately, improved the response to hypercapnia and hypoxia as soon as seven days after initiation of treatment (20), with normalization after two to five months of LT4 therapy only (20). The diminished diaphragmatic muscle strength normalized within three months of LT4 or LT3 treatment in both subclinical and overt hypothyroid patients (23 –25). As the evidence of an effect of hypothyroidism on the pulmonary function is too weak to draw any conclusions, the evidence of a direct effect of LT3 or LT4 substitution likewise becomes inadequate to allow any conclusions to be made on this topic (24,29 –31).
In hypothyroid patients with sleep apnea, most studies demonstrate an effect of LT4 therapy in reducing or even eliminating nocturnal apnea periods (33 –37).
No studies have investigated the treatment effects of LT3 compared to LT4 on respiratory symptoms, ventilation, pulmonary function, diaphragmatic function, or sleep apnea.
UAO and hypothyroidism
When hypothyroidism is caused by Hashimoto's thyroiditis, the effect of LT4 treatment can in part be explained by the reduction in thyroid size following restoration of euthyroidism (39,40). UAO does not always lead to subjective symptoms, and may therefore be overlooked in patients with thyroid disease (41). However, while thoroughly studied in nontoxic goiter (12), there are no studies on the pulmonary effect following goiter reduction in Hashimoto's thyroiditis.
The majority of Hashimoto's thyroiditis patients do not have a goiter. However, when present, a goiter can cause UAO by tracheal compression, as seen in 14–31% of patients referred for evaluation of simple goiter (41,42), and in 26–60% of patients referred to thyroidectomy (43 –45). Due to dislocation of the trachea or decrease of tracheal cross-sectional area, UAO can have a pronounced effect on the airflow (41,42,46), which is significantly improved by goiter volume reduction (47,48). Tracheal compression is somewhat more frequent when the goiter has a substernal location (35–73%), compared with a cervical location (9–58%) (49 –53). Tracheal compression might not lead to symptoms, but is nevertheless a relevant consideration, as UAO may develop into life-threatening respiratory insufficiency (41,54 –56). The extent to which these factors are at play in hypothyroid individuals is unknown.
Hypothyroidism and respiration in experimental animal models
Based on studies in experimental animal models, the brain stem, peripheral chemosensors, ATP generating enzymes, and muscle function, seem to be affected by hypothyroidism. Studies in hamsters—as in man (20)—have identified a decreased response (breathing frequency) to hypoxia and hypercapnia three months after the development of hypothyroidism (57 –59). This is caused by diminished dopamine receptor (D1) protein levels in the respiratory centers of the brain stem (paraventricular nucleus of the hypothalamus [PVN] and solitary nucleus) and the carotid glomus (57). Dopamine receptor (D2) levels are increased in the striatum and carotid glomus in the hypothyroid state, contrasting with the reduction of these receptor levels in the PVN (58).
A reduced diaphragmatic muscle strength, as seen in some human studies (23 –25), may be explained by studies in thyroidectomized rats (60). Here, a decreased capacity for energy transduction and glycolysis due to diminished enzyme levels (succinate dehydrogenase, hexokinase, 3-hydroxyl-CoA dehydrogenase, and phosphofructokinase) in the thoracic diaphragm has been found (60). Another study analyzing diaphragmatic muscle fibers from propylthiouracil-induced hypothyroidism in rats showed a decreased maximum force and myosin heavy chain2B/2X content (61).
From the aforementioned, there are many pathways to the diminished respiratory function, but it is clear that much remains to be explored in order to clarify how transient or permanent hypothyroidism, as well as LT4 treatment, affect the respiratory system.
Limitations
In the search for evidence of the impact of hypothyroidism on respiratory function, a paucity of high-quality data was encountered. No randomized controlled studies were conducted either in subclinical disease or in overt hypothyroidism in combination with medical treatment. Furthermore, many studies have a skewed balance between males and females, probably just reflecting the female preponderance of this disorder. Nevertheless, in addition to the small study populations, this questions whether the findings can be generalized to a broader population of hypothyroid individuals. Furthermore, many of the older studies do not address the potential confounding effect of the co-existence of goiter, if present.
Patients with pre-existing pulmonary disease were not included in this study. Therefore, it is not possible dissect whether hypothyroidism merely adds to the severity of pre-existing pulmonary disease or is the cause of de novo respiratory disease. There is some evidence that especially autoimmune hypothyroidism may be related to pulmonary diseases such as asthma and chronic obstructive pulmonary disease, but this area is sparsely investigated (62,63).
As hypothyroidism affects multiple organ systems, the distinction between a primary pulmonary effect and an effect mediated via the nervous system, the cardiopulmonary system, or the skeletal muscles is far from clear-cut. No data on cardiovascular or muscle function were included in the analyses, which might limit the generalizability of the findings. The impact of hypothyroidism on exercise capacity has been covered in a recent systematic review by Lankhaar et al. (64), who identified multiple causes of exercise intolerance due to disturbances in the cardiovascular, cardiopulmonary, musculoskeletal, neuromuscular, and cellular metabolic systems (64). Persisting complaints of exercise intolerance were found in a subgroup of subclinical hypothyroid patients not responding with symptom relief, despite adequate treatment with LT4 (64).
Implications for the future
Regarding the impact of overt hypothyroidism on respiratory symptoms, ventilation, and diaphragmatic muscle strength, no conclusions can be provided due to the small number of studies and the significant methodological weaknesses. Thus, this area of research is open for retesting hypotheses regarding the effects of hypothyroidism on various features of the respiratory system, employing prospective study designs, validated methods, and blinded assessments. The influence of overt hypothyroidism on the diaphragmatic function should also be further explored, since this unique thoraco-abdominal muscle has an important role in forming the voice and cough, and for exercise capacity. As for subclinical hypothyroidism, it remains uncertain whether the ventilator response to hypoxia and hypercapnia is affected, or whether these patients are more prone to suffer from OSAS than the background population.
Importantly, as most patients diagnosed with hypothyroidism are substituted with LT4, data on the long-term impact of persistent hypothyroidism on the respiratory function are virtually non-existent.
Conclusions
The evidence of an impact of hypothyroidism on respiratory function is at best limited. No information was found linking the effect of hypothyroidism on respiratory function to the increased mortality from pulmonary diseases, as reported previously from registry-based data (8). In contrast to the considerable knowledge regarding the influence of hypothyroidism on the cardiovascular system (65,66), many aspects of the influence on the respiratory function have been inadequately addressed or not explored at all.
Footnotes
Author Disclosure Statement
There are no conflicts of interest in this study.
