Exhaled breath metabolites reveal postmenopausal gut-bone cross-talk and non-invasive markers for osteoporosis

doi:10.21203/rs.3.rs-3996442/v1

Download PDF

Research Article

Exhaled breath metabolites reveal postmenopausal gut-bone cross-talk and non-invasive markers for osteoporosis

https://doi.org/10.21203/rs.3.rs-3996442/v1

This work is licensed under a CC BY 4.0 License

Version 1

posted

You are reading this latest preprint version

Menopause driven decline in estrogen exposes women to risk of osteoporosis. Detection of early onset and silent progression are keys to prevent fractures and associated burdens. In a discovery cohort of 120 postmenopausal women (aged: 49 – 90 years), we combined repeated quantitative pulse-echo ultrasonography of bone, assessment of grip strength and serum bone markers with mass-spectrometric analysis of exhaled metabolites to find breath volatile markers and quantitative cutoff levels for osteoporosis. Obtained markers and cutoffs were validated in an independent cohort of 49 age-matched women with 6 months apart seasonal follow-ups. Within the discovery cohort median concentrations (ppbV) of exhaled end-tidal dimethyl sulfide (DMS), allyl-methyl sulfide, butanethiol and butyric acid increased significantly (p≤0.005) by 45.24%, 42.83%, 38.29% and 33.66% in subjects (n=27) with bone density at high-risk of osteoporosis and fracture, when compared to subjects (n=62) with normal bone mineral density (BMD). Increased age and decreased grip strength were concomitant. All these changes were reproduced within the independent validation cohort including seasonal follow-ups. Exhaled metabolite expressions remained age independent in either cohort. Serum bone markers showed random expressions without any reproducibility. DMS exhalations even differed between patients with recent (<1 year), old and without fractures. Metabolite exhalations and BMDs were down-regulated during winter mirroring the regional climate change. ROC analysis in the discovery cohort yielded high classification accuracy (AUC=0.86) of DMS with a cutoff at 15.88 ppbV for osteoporosis, which predicted subjects at high-risk within the independent validation cohort with high (>91%) sensitivity and specificity during both follow-ups. Non-invasive analysis of exhaled DMS allowed more reliable classification of osteoporosis risk than conventional serum bone markers. We identified associations of exhaled organosulfur and short-chain fatty acid markers to bone metabolism in postmenopausal osteoporosis via a gut-bone axis.

Geriatrics & Gerontology

Physiology

Cellular Metabolism

Orthopedics

Translational Medicine

Anesthesiology & Pain Medicine

General Microbiology

Postmenopausal osteoporosis

Bone health

Aging

Breathomics

Bone markers

Breath biomarker

Volatile organic compound (VOC)

Metabolic monitoring

Gut-bone axis

Exhaled organosulfur compound offered >91% accurate option for non-invasive and personalized risk assessment of postmenopausal osteoporosis.

Bone is a highly dynamic tissue undergoing lifelong remodeling and serving endocrine, biochemical, and mechanical functions simultaneously (1). Both, bone turnover and mineral homeostasis are predominantly regulated by the interplay between bone cells and the endocrine system involving estrogen, parathormone, thyrocalcitonin, fibroblast growth factor 23 (FGF-23) and vitamin-D₃ (2). Menopause driven decline in estrogen exposes women to metabolic bone disorder leading towards rapid bone loss (3). Compromised bone health due to impaired bone mineralization and alterations of the extracellular matrix gradually turn into post-menopausal osteoporosis Type-1 – an advanced stage of compromised bone stability and a high susceptibility for bone fragility and fracture (4).

In general, onset and progression of postmenopausal osteoporosis is clinically silent and mainly remain unnoticed or minimally painful until the first low-energy fracture takes place (5). Roughly, one-third of world women population aged >50 years experiences osteoporotic fractures, out of which those of the distal radius (wrist) are often the first sign of osteoporosis and bone fragility (6). After the first fracture, women are at a two-fold increased risk to experience a second low-energy fracture within the next year. In women aged >45 years, osteoporosis accounts for longer periods of hospitalization than many other diseases, including diabetes, cardiovascular disease or breast cancer (7). This employs enormous clinical, economic and public health burdens globally – estimated €37 billion in Europe in 2010. Fracture managements represent >60% of such burden. Numbers of osteoporosis cases are expected to rise by >20% until 2025 in the EU (8). Thus, timely diagnosis is indispensable to administer early therapy and prevent fractures.

So far as clinical diagnosis is concerned, the dual-energy X-ray absorptiometry (DXA) scan for the determination of bone mineral density (BMD) is the WHO-suggested gold standard (9). Nevertheless, it exposes patients to X-ray radiation and is neither recommended to screen healthy or even asymptomatic individuals nor covered as a regular investigation by general health insurances. Eventually, FRAX questionnaires, bone echography, bone markers in blood and hand grip strength have been introduced to risk assessment (8, 10–12). While bone echography (e.g., BINDEX^® system based) exhibits 90% accuracy (with respect to the gold-standard DXA scan) to detect risk of osteoporosis and fracture in clinical studies, it requires skilled physicians for accurate measurements and critical interpretations to correctly categorize subjects (13, 14). Similarly, assessment of bone markers (e.g., BAP, TRAP5b, sclerostin, FGF-23 and Klotho etc.) are based on invasive blood sampling. Beyond this, there is no clear consensus on the selection of optimal markers and the interpretation of results are frequently confounded by intra-subject and/or intra- and inter-assay variations (15, 16). Though hand grip strength measurement is considered for musculoskeletal integrity assessment in osteoporosis (17), various confounding physio-metabolic effects from subject’s overall biological aging dilute a clear clinical relationship only to bone health (18, 19). Therefore, novel, inexpensive, non-invasive and repeatable diagnostic tools (with high-accuracy and reproducibility) are necessary for reliable risk assessments at the point-of-care and for personalized monitoring.

Various substance classes of endogenous volatile organic compounds (VOCs) are produced/regulated via diverse in vivo metabolic processes at the cellular and organ levels as well as by systemic microbiota (20). These VOCs are readily transported to the lung and exhaled through our breath. Real-time mass-spectrometry based breathomics offers rapid and repeated non-invasive assessment and continuous monitoring of various physiological (21–23), metabolic (24–26), pathological (27–29) and therapeutic (30–32) conditions/effects via exhaled VOCs. Years of development in real-time alveolar sampling and breath-resolved analysis allowed us to see minute systemic effects beyond the normal and confounding intra- and inter-individual physio-metabolic noise (33, 34). Nevertheless, well-designed prospective studies and independent validations are the prerequisites to define reliable breath markers and translate those into routine clinical practice for pathophysiological and therapeutic monitoring.

Here, in a real-life screening scenario, we report on the combination of bone echography, grip strength and invasive bone marker measurements with proton transfer reaction – time-of-flight – mass-spectrometric (PTR-ToF-MS) analysis of exhaled breath VOCs within a large discovery cohort and an age-matched independent validation cohort (with seasonal follow-ups) of postmenopausal women. Within the discovery cohort, we defined breath markers and quantitative cutoff values for osteoporosis risk assessment and applied those onto the validation cohort. Observed unprecedented associations between postmenopausal bone health and exhaled metabolic markers revealed gut-bone cross-talk and are reproduced within the independent validation cohort during repeated follow-ups. Our findings offered a new non-invasive option with high accuracy and translational potential for point-of-care and personalized monitoring of postmenopausal osteoporosis via breath biomarkers and opened up investigative avenues towards new therapeutic targets.

Recruited postmenopausal subjects, demography and lifestyle habits

Number (n=) of postmenopausal females recruited in discovery and independent validation cohorts from our bone-health screening campaign along with their bone-health status (based on BMD scores) are presented in Table 1.

Subject’s ethnicity, biological age, body mass index (BMI), habit of active smoking, consumption of alcohol (i.e., daily/ occasional), daily exercise, intake of special diet/ supplement and any other notable activities are also presented in Table 1. Due to the parametric distribution, age and BMI values are presented group wise as mean ± SD.

Clinical and therapeutic attributes

Number (n=) of subjects with old (>1 year from the participation date) and recent (≤1 year from the participation date) radius fractures and number (n=) of subjects undertaking bone-health related treatments (vitamin D₃, oral calcium, bisphosphonate, anti RNLK and hormone therapy) are listed in Table 1.

Bone, blood and breath markers

Quantitative values of BMD and blood markers for bone metabolism (BAP, TRAP5b, sclerostin, FGF23 and Klotho) are presented (parametric data) in terms of group wise mean ± SD in Table 1. Predominantly endogenous volatile metabolites in breath are also presented (non-parametric data) in terms of group wise median (minimum – maximum) in Table 1.

Relevant exhaled endogenous VOCs are comprised of organosulfur (dimethyl sulfide/DMS, allyl-methyl sulfide/AMS and butanethiol), short-chain fatty acid/SCFA (acetic acid, propionic acid and butyric acid), aldehyde (acetaldehyde, butyraldehyde, pentanal, hexanal), alcohol (ethanol), ketone (acetone) and hemiterpene (isoprene) substance classes.

Initial overview of measured parameters in both cohorts

BMD (normal, at risk and at high-risk of osteoporosis) wise categorization of measured parameters in both cohorts is presented as heatmap (Fig. 1). Relative differences in subjects’ BMD scores, biological age, hand strength, blood markers for bone metabolism and exhaled alveolar concentrations of volatile metabolites (with putative identification, protonated chemical formula and molar mass) are presented here. The heatmap represents non-quantitative expressions of each parameter normalized (independently in each cohort) onto the corresponding maximum value of an entire cohort. Here, both cohorts represent group wise means of normalized values from the initial recruitments (i.e., without seasonal follow-ups). Parameters are coded in different colors to depict relative differences.

Reproducibility of quantitative differences based on BMD or age

Subjects’ BMD and age wise categorizations of quantitative parameters differed significantly (p <0.05) and reproduced in both cohorts including seasonal follow-up (Fig. 2). All other parameters are presented in the supplement (fig. S1).

While categorized upon three groups of BMDs (left-side panels), statistically significant differences with respect to “BMD at high-risk” group are reproduced only in subjects’ biological age, grip strength (Fig. 2A) and exhaled alveolar concentrations of DMS, AMS, butanethiol and butyric acid (Fig. 2B).

While categorized upon three groups of biological age (right-side panels), apart from BMD scores and grip strength, all other significant differences (seen in BMD wise categorization) with respect to the age group “70 – 90 years” and their reproducibility disappeared (Fig. 2).

In all kind of categorizations, bone markers behaved at random and did not show any reproducibility (fig. S1).

Potential seasonal influence on BMDs and VOCs

While plotting all individual BMD scores (n= 358) from the discovery- (including repeated measures) and independent validation cohort (including seasonal follow-ups), as per the 12 calendar months of actual recruitments, those distributions closely mirrored the yearly profiles of regional sunlight and temperature exposures (fig. S2).

Within the independent validation cohort, exhalations of endogenous VOCs were down-regulated during the winter months (Fig. 2 and fig. S1).

Comparisons between subjects with and without radius fracture

Quantitative values of subjects’ biological age, hand strength and exhaled alveolar concentrations of DMS, AMS, butanethiol and butyric acid are presented based on the radius fracture status (no fractures, old fractures and recent fractures) and three groups BMD scores (Fig. 3). Within the “Recent fractures” cohort, subjects with “BMD at high-risk” range are severe cases of postmenopausal osteoporosis. Clinically relevant and statistically significant (p <0.05) differences with respect to the “high-risk” group are only observed in exhaled dimethyl sulfide concentrations.

Osteoporosis risk classification accuracy and independent validation including follow-up

Amongst all parameters, exhaled alveolar DMS concentrations yielded high AUC value of 0.86 (AUC bounds: 0.790 – 0.931 at 95% CI, standard error of 0.037 and asymptotic p <0.0001 under non-parametric assumptions) with a quantitative cutoff value of 15.88 ppbV for high-risk of osteoporosis in the discovery cohort.

Application of the discovery cohort derived quantitative cutoff onto the independent validation cohort resulted in >91% test accuracy (sensitivity of 91.7% and specificity of 91.3%) with respect to quantitative pulse-echo ultrasonography of bone (via Bindex^® device) in detecting subjects at high-risk of osteoporosis even during their seasonal follow-ups.

In this real-life observational screening study, we discovered and independently validated non-invasive breath VOC biomarkers for classifying postmenopausal osteoporosis. Combining bone ultrasonography, grip strength, bone marker measurements and PTR-ToF-MS analysis of exhaled breath VOCs within a large discovery cohort of postmenopausal women we identified breath markers and corresponding quantitative cutoffs for osteoporosis risk assessment. Applying these markers onto an age-matched independent cohort with seasonal follow-ups, we reproduced and validated the high accuracy of exhaled organosulfur compounds in addressing osteoporosis risk.

Reproducibility of demographic parameters as per BMD wise categorization denominates the strength of our real-life bone health screening campaign in both cohorts. As expected, subjects’ hand grip strength decreased with their compromised BMD and with increased age (17, 35) depicting the effect of aging on musculoskeletal health. Observed under-expressions of BMD scores and breath markers during the winter season indicate potential influences of sunlight and temperature on bone metabolism and thereby, justify the seasonal follow-up of the independent validation cohort. Surprisingly, in our study the anticipated seasonal effects (36) were absent on the serum bone markers. In routine clinical practice, a diagnostic marker must express differentially beyond the normal physio-metabolic limits as well as, intra- and inter-subject variations under the presence of a disease/condition. Although serum bone markers are repeatedly proposed for assessing the osteoporosis risk, we observed random expressions of these markers with relatively large variations and no tracible reproducibility in both BMD and age wise categorizations. This is in line with already reported confounding effects from intra-subject variations (15, 16) that hamper optimal bone marker selection in routine practice.

In contrast to invasive bone markers, non-invasive breath markers depicted close associations to postmenopausal bone health. Menopause driven estrogen decline causes rapid bone turnover in postmenopausal women whereas, a concomitant age-related bone loss remains slower and indifferent in both genders (4, 9). Reproducibility of exhaled organosulfur and SCFA expressions only in BMD wise categorizations indicates causal relationships of these VOCs to metabolic effects from postmenopausal bone health and not to the cumulative effects from subjects’ age.

Exhaled DMS, AMS, butanethiol and butyric acid are known to predominantly appear from the gut (37–39). There are evidences that a systemic microbiota-gut-bone axis regulates skeletal homeostasis by affecting host metabolism, gut permeability and nutrient absorption, cell mediated immunity, and endocrine functions (40–42). SCFAs (e.g., butyric acid) are produced via anaerobic microbial lysis of complex carbohydrates (e.g., undigested fibers/starch) and then diffuse from the gut lumen into the blood stream. Besides providing energy to the intestinal epithelial cells, promoting inflammatory response, enhancing gut-barrier integrity and Ca²⁺ and Mg²⁺ ions absorptions, butyric acid plays an important role in bone remodeling by directly promoting osteoblast activation and suppressing preosteoclast differentiation (41). While the links of SCFAs to bone health are repeatedly reported, such connections between bone metabolism and gut-related organosulfur compounds (e.g., DMS) are yet unknown and remained unexplored.

DMS is believed to be produced mainly during anaerobic microbial methylation of L-methionine in the lower gut (37, 43, 44). Earlier, we reported that AMS and butanethiol (/methyl-propyl sulfide) are largely produced from the oral and nasal cavity microbiota, respectively (45). Thus, unlike exhaled AMS and butanethiol profiles, DMS exhalations remained unaffected while switching between oral and nasal breathing. Therefore, the pronounced, significant and reproducible behavior of exhaled DMS with respect to systemic effects in our study, indicates its steady endogenous origin from the intestinal methionine metabolism.

Studies have also shown that dietary supplementation of methionine improves bone health and its long-term restriction deteriorate the same (46). In a pre-clinical model, methionine facilitated mRNA and protein expressions of important genes responsible for estrogen synthesis within primary ovarian granulosa cells (47). Methionine is a sulfur-containing essential amino acid that converts to homocysteine after donating the methyl group in cellular methylations (48). In vitro assays suggested that estrogen may regulate homocysteine levels through conjugation processes (49, 50). Although the systemic interactions between methionine/homocysteine and estrogen (its receptors) are complex and unclear (51), based on clinical investigations comparing pregnant, premenopausal and postmenopausal women with estrogen replacement therapy (ERT) with age-matched men and postmenopausal women without ERT, the 3^rd National Health and Nutrition Examination Survey has recognized the inverse relationship between serum estrogen and homocysteine (52). In line with that, randomized (placebo-)controlled trials also demonstrated that exogenous estrogen intervention in postmenopausal women inversely influences homocysteine–methionine metabolism (53, 54). However, vivid clinical investigations are lacking to address definite relationships between methionine metabolism, estrogen and bone health.

Previously, we demonstrated that in healthy premenopausal adults (without oral contraceptive pills), DMS exhalation tends to rise from the ovulation phase, soon after the estrogen level starts to decline physiologically. DMS exhalation decreases significantly and progressively in presence of daily oral contraceptives pills – containing synthetic estrogen and/or progesterone (25). While evaluating the physio-metabolic effects of healthy female aging, we observed that exhaled abundances of DMS tend to rise with age in postmenopausal cohorts (22). Such an ascending trend of DMS profiles with age is also noticed in the present study under age wise categorization and is observed significantly within the discovery cohort. These facts indicate that an increase (physiologically or administered artificially) in systemic estrogen level reduces in vivo DMS production in women.

As menopause driven progressive decline in physiological estrogen level gradually leads to progressive bone resorption (corresponding to bone loss rates of 3 – 5% within 5 – 10 years), significantly increased exhaled DMS concentrations in subjects at risk (/with osteopenia) and at high-risk of osteoporosis suggest a unique gut-bone metabolic axis – i.e., a potential and yet unexplored interplay between methionine and estrogen. Here, we hypothesize that DMS may act as a gut-microbial signaling molecule to actively regulate bone homeostasis under compromised BMD and, irrespective of an osteoporotic fracture event. Our hypothesis is further supported by the fact that DMS expressions did not differ significantly between “BMD at high-risk” groups based on their radius fracture status. We must remember that except for isoprene, the endogenous origin of any other breath VOC is rather putative/postulated. Our precedent multi-omic investigations has finally unraveled the true origin of isoprene in human breath (26) and thereby, dismissed the long-believed erroneous source (55) that has mislead clinical interpretations over the last four decades. Thus, our above hypothesis on DMS must be confirmed via further down-stream and system-wide investigations.

In order to focus explicitly on the physio-metabolic effects of osteoporosis, we excluded subjects with any severe acute/chronic comorbidities from the final analysis. Nevertheless, the age-related common comorbidities e.g., diabetes, hypertension and hypothyroidism could not be ruled out entirely. Noteworthy that our subjects with those comorbidities received corresponding medication/therapy and, therefore, were not experiencing any disease symptoms. They consented having a normal and healthy life.

In this context, breath acetone i.e., the putative byproduct of glycolysis and/or lipolysis (56) did not differ between groups in either cohort. Breath ethanol i.e., partly produced by microbial breakdown of carbohydrates in the small intestine (22), also remained indifferent. Similarly, breath aldehydes are regarded as markers for oxidative stress. Though, exhaled butanal and pentanal depicted a significant rise in the “high-risk” groups within the discovery cohort, reproducibility tests failed within the independent cohort. Being originated from the skeletal muscular lipolysis (26), exhaled isoprene remained unaffected by bone health status. Therefore, we could not consider any effect from well-treated comorbidities onto breath markers.

While looking at the limitations, the endogenous origins of observed breath markers are not strictly specific to human gut and/or its microbiome and are partly contributed due to genetic predisposition, systemic-microbial diversity, diet, lifestyle habits and geographical/environmental factors. Consequently, the outcomes and observed classification limits and quantitative cutoffs for gut-related VOCs may vary in another population/ ethnicity. Therefore, for generalizability, the identified marker(s) need to be reproduced in large independent screening studies on other ethnic populations. Further to that, our observations are restricted to postmenopausal bone health and cannot be generalized for all kind of osteoporosis especially, in male population. The state-of-the-art knowledge on the production/metabolism of VOCs in the gut and specific microbial colonies are largely based on in vitro and ex vivo observations. Longitudinal in vivo multi-omic investigations are indispensable to enhance our fundamental and mechanistic (from host and microbial genome towards exhaled volatile metabolome) understanding of individual/demographic group specific relationship and biochemical milieus between volatile markers and systemic microbiome under compromised bone health and/or other conditions/comorbidities of interest.

Observed associations between postmenopausal bone health and exhaled metabolic markers revealed certain unexplored avenues of gut-bone axis. Noteworthy that our study is not aimed to replace any well-established clinical diagnostic technique for osteoporosis. As postmenopausal osteoporosis progresses silently or with no/minimal symptoms, women do not recognize and/or ignore the underlying risk and do not go for a bone health checkup – until the first low-energy fracture occurs. Such a fracture in turn confirms the pronounced manifestation of osteoporosis. After the first fracture, the risk to experience a second fracture increases two fold, especially within the next year (7, 57). According to the EU guidelines for prevention and management of osteoporosis, physiological (or sufficient) level of 25(OH)D and calcium have to be achieved and the patients are encouraged to physical activity as a potent stimulus of bone formation. Beyond this, pharmacotherapy by means of an inhibition of osteoclast maturation and / or activation, i.e., the administration of bisphosphonates, human monoclonal antibody directed against RANKL, selective estrogen receptor modulators or a recombinant human PTH analogue are clinically used (58). These therapies may support the healing of an acute fracture and are directed to lower the risk for subsequent fractures. Nevertheless, the improvement of BMD and formation of new bone mass at the site of the fracture is time consuming. Consequently, those measures cannot completely cure an existing fracture in all cases (59). Thus, due to the delayed therapeutic intervention even before a low-energy fracture, bone health gradually deteriorates towards severe and irreversible osteoporosis. Here, our findings offered an unconventional non-invasive option with high test accuracy for osteoporosis risk classification and depicted translational potential for point-of-care and personalized monitoring of postmenopausal bone health and early onset of osteoporosis via breath biomarkers. In a perspective, our results also suggest investigative avenues towards new therapeutic targets e.g., via gut-microbiome specific pre-/probiotics.

Study design

This is a real-life observational screening study with independent prospective validation, conducted at Rostock University Medical Center, Rostock, Germany. The design contains a large discovery cohort of postmenopausal women (aged ≥49 years) and an age-matched independent validation cohort of women with repeated seasonal follow-ups. Within the discovery cohort, we identified breath markers and corresponding quantitative cutoff values for osteoporosis risk assessment and applied those onto the validation cohort including seasonal follow-ups.

Sample size estimation in both cohorts

Sample size in each cohort was calculated via analysis of variance (ANOVA) coupled with Bonferroni correction for multiple comparisons (between three groups of bone densities).

Given the high mass-resolving power of the PTR-ToF-MS used in this study, a minimal detectable difference in mean VOC intensities of 20 ncps (normalized counts pers second) allowed us to detect <5% difference in quantified trace (up to mid pptV range) concentrations of exhaled endogenous metabolites in both cohorts.

As per the real-life screening scenario and interindividual variations, we estimated a relatively high standard deviation of ± 30 ncps for the discovery cohort. Thus, in order to attain a test-power of 0.99 at 95% confidence interval (i.e., represented by a Bonferroni corrected alpha value of 0.016) the sample size of the discovery cohort resulted in 118 subjects.

Within the independent validation cohort, we applied the actual standard deviation of ±17 ncps obtained from the discovery cohort outcomes. Consequently, within the validation cohort, the alpha value of the discovery cohort was also readjusted further via Bonferroni correction and resulted as 0.005. Therefore, to achieve a test-power of 0.99 at 95% confidence interval the sample size in the independent validation cohort resulted in 45 subjects.

Recruitment of postmenopausal subjects

Ethical approvals (EA-No.: A 2017-0183 and A 2019-0040 for the discovery- and validation study, respectively) for clinical measurements were obtained from the institutional ethics committee (IEC) of Rostock University Medical Center (Germany) and all experiments were carried out in accordance with the amended Declaration of Helsinki guidelines.

All postmenopausal subjects were recruited from our clinical screening campaign of osteoporosis. Prior to participation, the study design and conduct were clearly explained to each participant by the principal investigator and a written and signed informed consent was obtained from each subject. These participants appeared to the clinic either by seeing the online advertisements of ongoing bone health screening campaign or due to a sudden radial fracture. Local travel costs of the participants were covered by institutional funds.

All recruitments and repeated (twice in a month) measurements within the discovery cohort took place from March until September 2018. The majority of these measurements were executed during the summer months (June – August 2018).

Initial recruitments in the independent validation cohort took place from October 2019 until March 2020. The majority of these initial measurements were executed during the winter months (December – February 2019). Thereafter, in order to abide by the mandatory safety regulations of the COVID-19 pandemic, we had to completely pause the follow-up measurements for 7 months. After resuming measurements during October – November 2020, these subjects were followed-up 6 months apart (during summer months of 2021) to observe any seasonal effect.

Exclusion criteria and measured parameters

In order to explicitly focus on bone health and to reduce extrinsic and intrinsic confounding factors, subjects with any severe acute or chronic comorbidity (excluding diabetes, hypothyroidism and hypertension), chronic/regular smoking and alcohol drinking habits, use of any special diet, supplement, medication (except oral medications for diabetes, hypothyroidism and hypertension) or therapy (except those with mandatory SARS-CoV-2 vaccinations and fracture treatments within the validation cohort) were excluded from final analysis. Subjects with abnormal respiratory rate (e.g., respiratory hyperventilation) were excluded. Similarly, in order to concentrate on early onset of the disease, subjects with immediate bone fracture other than in the radius were also excluded from data analysis.

Thus, we considered 120 subjects (out of 127 recruitments) from the discovery cohort and another 49 subjects (amongst 52 recruitments) from the independent validation cohort for actual data analysis and interpretations. These numbers also satisfied the above-estimated sample sizes.

After recording demography, lifestyle and clinical history related information, assessment of bone density index (via pulse-echo ultrasonography of bone) and grip strength were executed by experienced clinicians at the Dept. of Traumatology, Hand and Reconstructive Surgery. Within the next hour, real-time mass-spectrometric (PTR-ToF-MS) analysis of exhaled breath metabolites in these subjects were carried out at the Dept. of Anesthesiology, Intensive Care Medicine and Pain Therapy by an experienced medical scientist. Peripheral venous blood samples were collected for timely analysis of serum bone markers via commercially available biochemical assays at the Dept. of Pediatrics.

In case of follow-ups, approximately the same measurement time of the day was used for the same participant.

Assessment of bone density

The Bindex^® device (Bone Index Finland, Kuopio, Finland) was used for determination of bone density essentially as described previously (60). Briefly, measurements were taken at the distal radius and at two positions from the ipsilateral tibia. The ultrasound probe was placed at one third of radial length, i.e., the distance between the olecranon and the ulna styloid as well as one third of the tibial length below and above the top of the medial condyle and the medial malleolus, respectively. Ultrasound gel was applied and the probe was gently moved at the region of interest (ROI) and orthogonal to the surface of the underlying bone. Per ROI three valid signals were recorded. The proprietary software provided with the instrument (Bindex® software version 2.5) translated this information together with individual data on age, weight and height into a bone density [g/cm²] and a probability of osteoporosis (12).

As per age, gender and ethnicity-based recommendations of the Bindex® software, we considered BMD scores of >0.876 g/cm², 0.803 – 0.876 g/cm² and <0.803 g/cm²as normal, at risk (i.e., osteopenia) and at high-risk of osteoporosis, respectively.

Grip force monitoring

The Jamar® dynamometer (Type G200 from Biometrics Ltd., Newport, UK) was used for quantitative determination of grip force essentially as described (61). Briefly, participants were placed on a chair with the shoulder in a neutral position, the elbow 90° flexed and loosely gripping the handle of the dynamometer (62). Participants were asked to grab the dynamometer with maximum strength for 5 seconds, rest for 10 seconds and three repetitions per hand (61). The mean of three measurements was calculated and converted into Newton (i.e., equals to 0.224 pound).

Laboratory investigations

Blood was sampled in a plastic serum separator tube (Sarstedt, Nümbrecht, Germany), immediately transferred to the laboratory, allowed to clot at room temperature for approximately 20 min and centrifuged (2000 g, 15 min). Subsequently, samples were aliquoted and stored at -80°C until further analysis. In particular, the activities of the bone specific phosphatase (BAP) and tartrate-resistant acid phosphatase (TRAP5b) as well as the concentrations of sclerostin, soluble α-Klotho and intact fibroblast growth factor-23 (FGF23) were determined. To this purpose, assays for BAP and TRAP5b were purchased from IDS (Immunodiagnostic Systems Limited, Boldon Colliery, UK), the ones used for quantification of sclerostin, soluble α-Klotho and iFGF23 were obtained from TECO Medical Group (Sissach, Swiss), IBL (Immuno-Biological Laboratories, Minneapolis, USA) and Kainos Laboratories (Tokyo, Japan), respectively. All assays were used essentially as described and all samples were tested in duplicate (63, 64).

Breath sampling, VOC data analysis and quantification

Spontaneously breathing postmenopausal women maintained sitting posture (65) and performed oral breathing (45) via a customized mouthpiece (66) by following our state-of-the-are sampling procedure (67). Continuous side-stream sampling (flow: 20 mL/min) from the mouthpiece was performed via the heated (75 °C) transfer-line of a PTR-ToF-MS-8000 (Ionicon Analytik GmbH, Innsbruck, Austria) under optimized conditions (68, 69). We used a PTR time-resolution of 200 ms, drift-tube temperature of 75 °C, pressure of 2.3 mbar and voltage of 610 V in order to reach the E/N ratio of 139 Td (22, 27, 55). The mass scale was recalibrated automatically based on three masses viz., 21.0226 (H₃O⁺-isotope), 29.998 (NO⁺) and 59.049 (protonated C₃H₆O) after each minute of data accusation.

A PTR-MS viewer software (version 3.228) was used for raw data processing. VOCs were measured continuously in counts per second (cps), which were normalized onto the corresponding counts (i.e., to obtain VOC data in ncps) of the primary ion (H₃O⁺). Breath-by-breath assignment of inspired (room air) and expired (alveolar/end-tidal) phases of breath were performed via custom-made ‘breath tracker’ algorithm (70, 71). Here, an endogenous VOC (e.g., acetone) with orders of magnitudes higher concentration in exhalation than in room air was used as the tracker mass.

We quantified the VOCs either via reaction rate constant (k-rates) between the volatile and H₃O⁺ ion (at the E/N ratio of 140 Td) or via multi-component VOC standard mixture under breath adapted humidity and CO₂ conditions by using a liquid calibration unit (LCU, Ionicon Analytik GmbH, Innsbruck, Austria)(72).

Statistical analysis

For statistical comparisons of bone health, postmenopausal subjects in either cohort were categorized separately upon BMD scores and biological age. BMD wise categorization of subjects was based on normal (>0.876 g/cm²), at risk (0.803 – 0.876 g/cm²) and at high-risk (<0.803 g/cm²) ranges of bone density scores. Age wise distributions of subjects were based on 49 – 59 years, 60 – 69 years and 70 – 90 years.

In order to observe any effect of fractures, subjects from both cohorts with old fractures (i.e., occurring >1 year before the recruitment), recent radius fracture (i.e., occurring within 12 months prior to enrolment) and without a history of fractures were categorized upon three levels of BMD sores.

In order to check the type of distribution, either by a Kolmogorov-Smirnov test (where N ≥50) or a Shapiro-Wilk (where N <50) test for normality (that failed at p <0.050) was applied.

In case of parametric data, we used mean values and statistically significant differences between study groups were tested independently within each cohort by means of one-way ANOVA test followed by a pairwise multiple comparisons test via post-hoc Dunn’s method (at p <0.05) due to unequal group sizes.

In case of non-parametric data, we used median values and statistically significant differences between study groups were tested independently within each cohort by means of Kruskal-Willis H test (i.e., a one-way ANOVA on Ranks test for non-parametric data) followed by the pairwise multiple comparisons test via post-hoc Dunn’s method (at p <0.05) due to unequal group sizes.

Receiver operating characteristic (ROC) curves were generated from the discovery cohorts, based on subject’s biological age, grip strength, bone markers and exhaled alveolar concentrations of dimethyl sulfide, allyl-methyl sulfide, butanethiol, and butyric acid. Area under the curve (AUC) and cutoff values were calculated with corresponding standard error and statistical significance (p <0.05).

Discovery cohort derived cutoff values of parameter(s) with high classification accuracy (where AUC value was >0.85, standard error was <5% and asymptotic significance level was p <0.005) were applied to predict the subjects with “BMD at high-risk” of osteoporosis within the validation cohort including follow-ups. Resulting ROC curves with corresponding test sensitivity and specificity are presented.

Acknowledgments:

We thank all participants for their voluntary participation in this study.

Funding:

European Union fund for regional development EFRE (JKS, WM)

EU Horizon-2020 grant H2020-PCH-HEARTEN project 643694 (WM, JKS)

European Union’s Horizon 2020 Marie Skłodowska-Curie research and innovation programme 674911-IMPACT (JKS, WM)

University Medicine Rostock’s FORUN programme (2018) 889003 (PS)

Author contributions:

Conceptualization: PS, DCF

Study design: PS, DCF, TM, JKS, WM

Methodology: PS, DCF, TM, DAR, BK, JKS, WM

Investigation: CB, SG, PS

Analysis: PS, CB, SG, AR

Visualization: PS, DCF, DAR, BK, WM

Interpretation: PS, DCF, TM, BK, DAR, JKS, WM

Funding acquisition: PS, DCF, TM, DAR, JKS, WM

Project administration: PS, DCF, TM, JKS, WM

Supervision: PS, DCF, TM, JKS, WM

Writing – original draft: PS

Writing – review & editing: DCF, CB, SG, AR, TM, BK, DAR, JKS, WM, PS

Competing interests:

Authors declare that they have no competing interests.

Data and materials availability:

All disclosable data are available in the main text or the supplementary materials. Raw and processed experimental data is available from the corresponding author upon reasonable request.

S. Harada, G. A. Rodan, Control of osteoblast function and regulation of bone mass. Nature 423, 349–355 (2003).
S. Fukumoto, T. J. Martin, Bone as an endocrine organ. Trends Endocrinol. Metab. 20, 230–236 (2009).
S. L. Teitelbaum, Bone Resorption by Osteoclasts. Science 289, 1504–1508 (2000).
R. Eastell, T. W. O’Neill, L. C. Hofbauer, B. Langdahl, I. R. Reid, D. T. Gold, S. R. Cummings, Postmenopausal osteoporosis. Nature Reviews Disease Primers 2, nrdp201669 (2016).
S. Alami, L. Hervouet, S. Poiraudeau, K. Briot, C. Roux, Barriers to Effective Postmenopausal Osteoporosis Treatment: A Qualitative Study of Patients’ and Practitioners’ Views. PLoS One 11 (2016), doi:10.1371/journal.pone.0158365.
T. Sözen, L. Özışık, N. Ç. Başaran, An overview and management of osteoporosis. Eur J Rheumatol 4, 46–56 (2017).
J. A. Kanis, C. Cooper, R. Rizzoli, B. Abrahamsen, N. M. Al-Daghri, M. L. Brandi, J. Cannata-Andia, B. Cortet, H. P. Dimai, S. Ferrari, P. Hadji, N. C. Harvey, M. Kraenzlin, A. Kurth, E. McCloskey, S. Minisola, T. Thomas, J.-Y. Reginster, Identification and management of patients at increased risk of osteoporotic fracture: outcomes of an ESCEO expert consensus meeting. Osteoporos Int 28, 2023–2034 (2017).
J. A. Kanis, E. V. McCloskey, H. Johansson, C. Cooper, R. Rizzoli, J.-Y. Reginster, Scientific Advisory Board of the European Society for Clinical and Economic Aspects of Osteoporosis and Osteoarthritis (ESCEO) and the Committee of Scientific Advisors of the International Osteoporosis Foundation (IOF), European guidance for the diagnosis and management of osteoporosis in postmenopausal women. Osteoporos Int 24, 23–57 (2013).
M. D. Walker, E. Shane, Postmenopausal Osteoporosis. New England Journal of Medicine 389, 1979–1991 (2023).
R. H. Christenson, Biochemical Markers of Bone Metabolism: An Overview. Clinical Biochemistry 30, 573–593 (1997).
W. G. Dixon, M. Lunt, S. R. Pye, J. Reeve, D. Felsenberg, A. J. Silman, T. W. O’Neill, on behalf of the European Prospective Osteoporosis Study Group, Low grip strength is associated with bone mineral density and vertebral fracture in women. Rheumatology 44, 642–646 (2005).
J. P. Karjalainen, O. Riekkinen, J. Töyräs, J. S. Jurvelin, H. Kröger, New method for point-of-care osteoporosis screening and diagnostics. Osteoporos Int 27, 971–977 (2016).
J. T. Schousboe, O. Riekkinen, J. Karjalainen, Prediction of hip osteoporosis by DXA using a novel pulse-echo ultrasound device. Osteoporos Int 28, 85–93 (2017).
E. M. Lewiecki, Pulse-echo Ultrasound Identifies Caucasian and Hispanic Women at Risk for Osteoporosis. Journal of Clinical Densitometry 24, 175–182 (2021).
X. Filella, N. Guañabens, Clinical use of bone markers: a challenge to variability. Advances in Laboratory Medicine / Avances en Medicina de Laboratorio (2023), doi:10.1515/almed-2023-0092.
L. Koch, Assay variability of biochemical markers of bone turnover. Nat Rev Endocrinol 5, 587–587 (2009).
J. Pratt, G. De Vito, M. Narici, R. Segurado, J. Dolan, J. Conroy, C. Boreham, Grip strength performance from 9431 participants of the GenoFit study: normative data and associated factors. GeroScience 43, 2533–2546 (2021).
J. K. Richardson, T. J. Ellmers, The relationship between clinical measures of cognitive function and grip strength in healthy older adults. BMC Geriatr 22, 907 (2022).
R. G. Carson, Get a grip: individual variations in grip strength are a marker of brain health. Neurobiology of Aging 71, 189–222 (2018).
W. Miekisch, P. Sukul, J. K. Schubert, in Volatile Biomarkers for Human Health: From Nature to Artificial Senses, (The Royal Society of Chemistry, 2023), pp. 22–38.
P. Sukul, J. Bartels, P. Fuchs, P. Trefz, R. Remy, L. Rührmund, S. Kamysek, J. K. Schubert, W. Miekisch, Effects of COVID-19 protective face-masks and wearing durations onto respiratory-haemodynamic physiology and exhaled breath constituents. European Respiratory Journal 60, 2200009 (2022).
P. Sukul, S. Grzegorzewski, C. Broderius, P. Trefz, T. Mittlmeier, D.-C. Fischer, W. Miekisch, J. K. Schubert, Physiological and metabolic effects of healthy female aging on exhaled breath biomarkers. iScience 25, 103739 (2022).
P. Mochalski, J. King, C. A. Mayhew, K. Unterkofler, A review on isoprene in human breath. J. Breath Res. (2023), doi:10.1088/1752-7163/acc964.
G. Pugliese, P. Trefz, M. Weippert, J. Pollex, S. Bruhn, J. K. Schubert, W. Miekisch, P. Sukul, Real-time metabolic monitoring under exhaustive exercise and evaluation of ventilatory threshold by breathomics: Independent validation of evidence and advances. Frontiers in Physiology 13 (2022) (available at https://www.frontiersin.org/articles/10.3389/fphys.2022.946401).
P. Sukul, J. K. Schubert, P. Trefz, W. Miekisch, Natural menstrual rhythm and oral contraception diversely affect exhaled breath compositions. Scientific Reports 8, 10838 (2018).
P. Sukul, A. Richter, C. Junghanss, J. K. Schubert, W. Miekisch, Origin of breath isoprene in humans is revealed via multi-omic investigations. Commun Biol 6, 1–12 (2023).
R. Remy, N. Kemnitz, P. Trefz, P. Fuchs, J. Bartels, A.-C. Klemenz, L. Rührmund, P. Sukul, W. Miekisch, J. K. Schubert, Profiling of exhaled volatile organics in the screening scenario of a COVID-19 test center. iScience , 105195 (2022).
W. Ibrahim, M. J. Wilde, R. L. Cordell, M. Richardson, D. Salman, R. C. Free, B. Zhao, A. Singapuri, B. Hargadon, E. A. Gaillard, T. Suzuki, L. L. Ng, T. Coats, P. Thomas, P. S. Monks, C. E. Brightling, N. J. Greening, S. Siddiqui, ON BEHALF OF THE EMBER CONSORTIUM, Visualization of exhaled breath metabolites reveals distinct diagnostic signatures for acute cardiorespiratory breathlessness. Science Translational Medicine 14, eabl5849 (2022).
P. Trefz, S. C. Schmidt, P. Sukul, J. K. Schubert, W. Miekisch, D.-C. Fischer, Non-Invasive Assessment of Metabolic Adaptation in Paediatric Patients Suffering from Type 1 Diabetes Mellitus. J Clin Med 8 (2019), doi:10.3390/jcm8111797.
B. Löser, A. Grabenschröer, G. Pugliese, P. Sukul, P. Trefz, J. K. Schubert, W. Miekisch, Changes of Exhaled Volatile Organic Compounds in Postoperative Patients Undergoing Analgesic Treatment: A Prospective Observational Study. Metabolites 10, 321 (2020).
P. Trefz, S. Kamysek, P. Fuchs, P. Sukul, J. K. Schubert, W. Miekisch, Drug detection in breath: non-invasive assessment of illicit or pharmaceutical drugs. J. Breath Res. 11, 024001 (2017).
B. Brock, S. Kamysek, J. Silz, P. Trefz, J. K. Schubert, W. Miekisch, Monitoring of breath VOCs and electrical impedance tomography under pulmonary recruitment in mechanically ventilated patients. J. Breath Res. 11, 016005 (2017).
P. Sukul, P. Trefz, in Bioanalytical Reviews. (Springer, Berlin, Heidelberg, 2022), vol. 4, pp. 1–18.
A. S. Modak, Why have only a handful of breath tests made the transition from R&D to clinical practice? J. Breath Res. 18, 012001 (2023).
J. Sirola, M. Tuppurainen, R. Honkanen, J. S. Jurvelin, H. Kröger, Associations between grip strength change and axial postmenopausal bone loss—a 10-year population-based follow-up study. Osteoporos Int 16, 1841–1848 (2005).
P. B. Rapuri, H. K. Kinyamu, J. C. Gallagher, V. Haynatzka, Seasonal Changes in Calciotropic Hormones, Bone Markers, and Bone Mineral Density in Elderly Women. The Journal of Clinical Endocrinology & Metabolism 87, 2024–2032 (2002).
A. Tangerman, Measurement and biological significance of the volatile sulfur compounds hydrogen sulfide, methanethiol and dimethyl sulfide in various biological matrices. Journal of Chromatography B 877, 3366–3377 (2009).
Y. P. Silva, A. Bernardi, R. L. Frozza, The Role of Short-Chain Fatty Acids From Gut Microbiota in Gut-Brain Communication. Front. Endocrinol. 11 (2020), doi:10.3389/fendo.2020.00025.
L. Wang, S. Cen, G. Wang, Y. Lee, J. Zhao, H. Zhang, W. Chen, Acetic acid and butyric acid released in large intestine play different roles in the alleviation of constipation. Journal of Functional Foods 69, 103953 (2020).
M. M. Zaiss, R. M. Jones, G. Schett, R. Pacifici, The gut-bone axis: how bacterial metabolites bridge the distance. J Clin Invest 129, 3018–3028 (2019).
Y. Tu, R. Yang, X. Xu, X. Zhou, The microbiota-gut-bone axis and bone health. Journal of Leukocyte Biology 110, 525–537 (2021).
Y. He, Y. Chen, The potential mechanism of the microbiota-gut-bone axis in osteoporosis: a review. Osteoporos Int 33, 2495–2506 (2022).
K. Kalantar-Zadeh, K. J. Berean, R. E. Burgell, J. G. Muir, P. R. Gibson, Intestinal gases: influence on gut disorders and the role of dietary manipulations. Nat Rev Gastroenterol Hepatol 16, 733–747 (2019).
E. Mutuyemungu, M. Singh, S. Liu, D. J. Rose, Intestinal gas production by the gut microbiota: A review. Journal of Functional Foods 100, 105367 (2023).
P. Sukul, P. Oertel, S. Kamysek, P. Trefz, Oral or nasal breathing? Real-time effects of switching sampling route onto exhaled VOC concentrations. J. Breath Res. 11, 027101 (2017).
U. Navik, V. G. Sheth, A. Khurana, S. S. Jawalekar, P. Allawadhi, R. R. Gaddam, J. S. Bhatti, K. Tikoo, Methionine as a double-edged sword in health and disease: Current perspective and future challenges. Ageing Research Reviews 72, 101500 (2021).
G. Yang, S. Li, S. Cai, J. Zhou, Q. Ye, S. Zhang, F. Chen, F. Wang, X. Zeng, Dietary methionine supplementation during the estrous cycle improves follicular development and estrogen synthesis in rats. Food & Function 15, 704–715 (2024).
A. M. Troen, E. Lutgens, D. E. Smith, I. H. Rosenberg, J. Selhub, The atherogenic effect of excess methionine intake. Proc Natl Acad Sci U S A 100, 15089–15094 (2003).
S. Shah, R. J. Bell, S. R. Davis, Homocysteine, estrogen and cognitive decline. Climacteric 9, 77–87 (2006).
N. W. Gaikwad, Mass spectrometry evidence for formation of estrogen–homocysteine conjugates: Estrogens can regulate homocysteine levels. Free Radical Biology and Medicine 65, 1447–1454 (2013).
K. R. Dimitrova, K. DeGroot, A. K. Myers, Y. D. Kim, Estrogen and homocysteine. Cardiovascular Research 53, 577–588 (2002).
M. S. Morris, P. F. Jacques, J. Selhub, I. H. Rosenberg, Total Homocysteine and Estrogen Status Indicators in the Third National Health and Nutrition Examination Survey. American Journal of Epidemiology 152, 140–148 (2000).
R. G. V. Smolders, K. de Meer, P. Kenemans, T. Teerlink, C. Jakobs, M. J. van der Mooren, Hormone replacement influences homocysteine levels in the methionine-loading test: a randomized placebo controlled trial in postmenopausal women. Eur J Obstet Gynecol Reprod Biol 117, 55–59 (2004).
A. Cagnacci, M. Generali, D. Pirillo, F. Baldassari, A. Volpe, Effects of low- or high-dose hormone therapy on fasting and post-methionine homocysteine levels in postmenopausal women. Climacteric 9, 388–395 (2006).
P. Sukul, A. Richter, J. K. Schubert, W. Miekisch, Deficiency and absence of endogenous isoprene in adults, disqualified its putative origin. Heliyon 7, e05922 (2021).
M. P. Kalapos, On the mammalian acetone metabolism: from chemistry to clinical implications. Biochim. Biophys. Acta 1621, 122–139 (2003).
A. R. Cappola, D. M. Shoback, Osteoporosis Therapy in Postmenopausal Women With High Risk of Fracture. JAMA 316, 715–716 (2016).
J. A. Kanis, C. Cooper, R. Rizzoli, J.-Y. Reginster, European guidance for the diagnosis and management of osteoporosis in postmenopausal women. Osteoporos Int 30, 3–44 (2019).
L. M. McNamara, Perspective on post-menopausal osteoporosis: establishing an interdisciplinary understanding of the sequence of events from the molecular level to whole bone fractures. J R Soc Interface 7, 353–372 (2010).
M. Behrens, S. Felser, A. Mau-Moeller, M. Weippert, J. Pollex, R. Skripitz, P. K. E. Herlyn, D.-C. Fischer, S. Bruhn, H.-C. Schober, V. Zschorlich, T. Mittlmeier, The Bindex® ultrasound device: reliability of cortical bone thickness measures and their relationship to regional bone mineral density. Physiol. Meas. 37, 1528 (2016).
M. Mühldorfer-Fodor, S. Ziegler, C. Harms, J. Neumann, A. Cristalli, A. Kalpen, G. Kundt, T. Mittlmeier, K. J. Prommersberger, Grip force monitoring on the hand: Manugraphy system versus Jamar dynamometer. Arch Orthop Trauma Surg 134, 1179–1188 (2014).
V. Mathiowetz, K. Weber, G. Volland, N. Kashman, Reliability and validity of grip and pinch strength evaluations. The Journal of Hand Surgery 9, 222–226 (1984).
A. Doyon, D.-C. Fischer, A. K. Bayazit, N. Canpolat, A. Duzova, B. Sözeri, J. Bacchetta, A. Balat, A. Büscher, C. Candan, N. Cakar, O. Donmez, J. Dusek, M. Heckel, G. Klaus, S. Mir, G. Özcelik, L. Sever, R. Shroff, E. Vidal, E. Wühl, M. Gondan, A. Melk, U. Querfeld, D. Haffner, F. Schaefer, 4C Study Consortium, Markers of bone metabolism are affected by renal function and growth hormone therapy in children with chronic kidney disease. PLoS ONE 10, e0113482 (2015).
D.-C. Fischer, C. Smith, F. De Zan, J. Bacchetta, S. A. Bakkaloglu, A. Agbas, A. Anarat, B. Aoun, V. Askiti, K. Azukaitis, A. Bayazit, I. K. Bulut, N. Canpolat, D. Borzych-Dużałka, A. Duzova, S. Habbig, S. Krid, C. Licht, M. Litwin, L. Obrycki, F. Paglialonga, A. Rahn, B. Ranchin, C. Samaille, M. Shenoy, M. D. Sinha, B. Spasojevic, C. J. Stefanidis, E. Vidal, A. Yilmaz, M. Fischbach, F. Schaefer, C. P. Schmitt, R. Shroff, Hemodiafiltration Is Associated With Reduced Inflammation and Increased Bone Formation Compared With Conventional Hemodialysis in Children: The HDF, Hearts and Heights (3H) Study. Kidney International Reports 6, 2358–2370 (2021).
P. Sukul, P. Trefz, S. Kamysek, J. K. Schubert, W. Miekisch, Instant effects of changing body positions on compositions of exhaled breath. J. Breath Res. 9, 047105 (2015).
P. Sukul, J. K. Schubert, S. Kamysek, P. Trefz, W. Miekisch, Applied upper-airway resistance instantly affects breath components: a unique insight into pulmonary medicine. J. Breath Res. 11, 047108 (2017).
P. Sukul, J. K. Schubert, K. Zanaty, P. Trefz, A. Sinha, S. Kamysek, W. Miekisch, Exhaled breath compositions under varying respiratory rhythms reflects ventilatory variations: translating breathomics towards respiratory medicine. Scientific Reports 10, 14109 (2020).
P. Sukul, P. Trefz, J. K. Schubert, W. Miekisch, Advanced setup for safe breath sampling and patient monitoring under highly infectious conditions in the clinical environment. Sci Rep 12, 17926 (2022).
P. Trefz, M. Schmidt, P. Oertel, J. Obermeier, B. Brock, S. Kamysek, J. Dunkl, R. Zimmermann, J. K. Schubert, W. Miekisch, Continuous real time breath gas monitoring in the clinical environment by proton-transfer-reaction-time-of-flight-mass spectrometry. Anal. Chem. 85, 10321–10329 (2013).
H. Schwoebel, R. Schubert, M. Sklorz, S. Kischkel, R. Zimmermann, J. K. Schubert, W. Miekisch, Phase-resolved real-time breath analysis during exercise by means of smart processing of PTR-MS data. Anal Bioanal Chem 401, 2079–2091 (2011).
P. Sukul, P. Trefz, J. K. Schubert, W. Miekisch, Immediate effects of breath holding maneuvers onto composition of exhaled breath. J Breath Res 8, 037102 (2014).
P. Trefz, J. K. Schubert, W. Miekisch, Effects of humidity, CO2 and O2 on real-time quantitation of breath biomarkers by means of PTR-ToF-MS. J Breath Res 12, 026016 (2018).

Table 1

Demography, lifestyle, clinical status and ongoing therapy.
		Discovery cohort				Independent validation cohort
Screening period		March – September 2018 (mainly: Summer)				October – March 2019 (mainly: Winter)
	Postmenopausal females	Total	Females at normal bone health	Females at risk / Osteopenia	Females at high-risk / Osteoporosis	Total	Females at normal bone health	Females at risk / Osteopenia	Females at high-risk / Osteoporosis
Number (n =)		120	62	31	27	49	33	10	6
Demography
	Ethnicity / Region	Caucasian / European (mainly: German)				Caucasian / European (mainly: German)
	Age (years)	65.3 ± 8.1	61.5 ± 7.0	67.5 ± 7.4	71.4 ± 6.9	63.3 ± 8.9	60.5 ± 7.6	67.4 ± 8.1	71.5 ± 10.2
	BMI (kg/m²)	25.2 ± 4.5	25.9 ± 4.7	25.2 ± 4.4	23.6 ± 3.7	24.1 ± 4.1	25.0 ± 4.2	23.2 ± 3.8	21.3 ± 2.1
Lifestyle habits and clinical status
	Active smoking (n =)	13	7	2	4	3	3	0	0
	Alcohol consumption (n =)	11	7 occasional	4 occasional	0	4	4 occasional	0	0
	Daily exercise (n =)	9	6	3	0	4	3	1	0
	Other notable activities	None				None
	Acute/chronic comorbidity*	None				None
	Other notable condition	None				SARS-CoV-2 infection (at initial = 0, until final follow-up = 27)
	Other notable attributes	None				SARS-CoV-2 vaccination (at initial = 0, until final follow-up = 49)
Bone health related attributes
	Vitamin D₃ (n =)	17	3	6	8	24	16	4	4
	Oral calcium (n =)	7	1	3	2	2	1	1	0
	Bisphosphonate (n =)	10	2	6	7	1	0	1	0
	Hormone (n =)	1	1	0	0	2	1	1	0
	anti RANLK (n =)	1	0	0	1	1	1	0	0
	Radius fracture (n =)	39	13	12	14	11	6	3	2
	Old fracture (n =)	21	7	7	7	3	2	1	0
	Recent fracture (n =)	18	6	5	7	8	5	1	2
Measured clinical parameters from bone, hand strength and blood
	BMD (g/cm²)**	0.882 ± 0.10	0.959 ± 0.07	0.844 ± 0.02	0.745 ± 0.04	0.902 ± 0.09	0.953 ± 0.05	0.835 ± 0.02	0.734 ± 0.01
	Grip strength (N)	234.6 ± 49.9	249.8 ± 43.2	230.4 ± 53.9	204.5 ± 46.52	245.5 ± 63.53	260.7 ± 53.2	247.4 ± 51.1	158.6 ± 72.89
	BAP (U/L)	20.4 ± 6.55	19.0 ± 6.04	20.6 ± 5.68)	23.3 ± 7.77	15.6 ± 4.41	15.2 ± 4.25	14.7 ± 5.14	19.0 ± 2.78
	TRAP5b (U/L)	3.38 ± 1.30	3.36 ± 1.35	3.25 ± 1.10	3.54 ± 1.42	3.02 ± 1.13	2.87 ± 1.10	3.26 ± 1.14	3.50 ± 1.10
	Sclerostin (ng/mL)	0.37 ± 0.14	0.38 ± 0.15	0.37 ± 0.13	0.34 ± 0.16	0.65 ± 0.22	0.72 ± 0.20	0.40 ± 0.17	0.68 ± 0.18
	FGF23 (pg/mL)	38.13 ± 20.1	37.23 ± 21.0	40.60 ± 20.4	37.43 ± 17.90	45.68 ± 16.05	46.50 ± 15.7	42.38 ± 14.1	46.83 ± 22.39
	Klotho (pg/mL)	835 ± 256.1	846.7 ± 292	828.2 ± 267	815.8 ± 127	785.4 ± 202.8	732.0 ± 167	915.2 ± 241	884.5 ± 228.0
Potentially endogenous volatile metabolites in exhaled breath
	Dimethyl sulfide (ppbV)	14.8 (3.20–77.4)	12.9 (3.20–30.8)	11.9 (6.40–36.9)	23.6 (12.3–77.4)	8.70 (3.00–38.6)	6.80 (3.00–13.1)	10.7 (6.90–25.4)	25.6 (13.1–38.6)
	Allyl-methyl sulfide (ppbV)	4.30 (1.30–27.3)	4.21 (1.30–25.9)	3.65 (1.30–21.5)	7.36 (1.80–27.3)	2.30 (0.40–17.7)	1.78 (0.50–8.10)	2.47 (0.40–17.7)	9.32 (1.40–17.7)
	Butanethiol (ppbV)	1.21 (0.48–16.4)	1.15 (0.48–6.81)	1.13 (0.55–5.63)	1.74 (0.76–16.4)	0.50 (0.23–1.97)	0.45 (0.27–1.97)	0.52 (0.23–1.11)	1.31 (0.40–1.66)
	Acetic acid (ppbV)	58.7 (15.8–303)	56.4 (15.9–212)	55.3 (26.1–203)	80.8 (15.8–303)	34.0 (8.20–165)	35.5 (9.20–95.4)	34.7 (8.20–108)	33.8 (8.60–165)
	Propanoic acid (ppbV)	20.5 (4.60–143)	17.3 (5.60–122)	18.8 (8.20–141)	25.4 (4.60–143)	8.60 (2.00–60.6)	7.50 (2.40–60.6)	10.6 (1.90–42.9)	11.9 (3.50–47.8)
	Butyric acid (ppbV)	4.50 (1.10–52.7)	4.20 (1.10–32.6)	4.10 (1.30–24.0)	6.90 (1.20–52.7)	2.10 (0.50–15.6)	2.10 (0.60–9.4)	1.90 (0.50–15.6)	6.10 (1.30–10.7)
	Acetaldehyde (ppbV)	243.2 (125.6–648.6)	237.9 (125.6–466.1)	234.5 (134.7–628.4)	262.9 (136.7–648.6)	225.5 (107.3–653.9)	206.7 (107.3–653.9)	250.3 (145.8–500.5)	211.5 (142.8–427.0)
	Butyraldehyde (ppbV)	14.5 (2.80–93.0)	13.0 (2.80–45.8)	12.8 (2.90–93.0)	20.2 (5.80–74.5)	4.50 (0.60–138.3)	4.50 (1.30–73.3)	4.50 (0.60–138.3)	4.60 (2.40–8.2)
	Pentanal (ppbV)	10.9 (4.30–29.5)	10.7 (4.40–18.6)	10.6 (4.30–17.1)	14.1 (7.10–29.5)	7.70 (3.50–31.9)	7.70 (3.50–31.9)	7.90 (4.00–26.0)	6.70 (5.10–8.00)
	Hexanal (ppbV)	2.8 (0.8–8.0)	2.7 (0.8–7.0)	2.6 (0.8–5.5)	3.2 (1.6–8.0)	2.7 (0.1–10)	3.2 (0.1–10)	2.9 (0.2–4.9)	1.8 (0.7–2.9)
	Ethanol (ppbV)	105.3 (4.40–1160)	92.5 (4.40–767.0)	105.2 (9.70–496.0)	255.5 (49.6–1160)	38.80 (8.00–646.3)	30.80 (8.00–528.9)	60.80 (12.00–646.3)	52.50 (30.70–421.0)
	Acetone (ppbV)	216.9 (154.9–292.0)	219.2 (154.9–292.0)	211.5 (154.9–253.0)	216.6 (186.5–268.3)	199.0 (125.9–286.5)	199.0 (125.9–266.3)	193.9 (154.3–275.2)	231.1 (174.6–286.5)
	Isoprene (ppbV)	98.40 (27.90–226.9)	118.2 (46.40–226.9)	112.5 (70.30–185.8)	61.50 (27.90–116.0)	57.20 (9.90–127.5)	51.30 (9.90–102.5)	59.50 (10.70–104.0)	69.10 (26.60–127.5)
Parametric variables are presented as mean ± SD values. Non-parametric variables are presented as median (minimum – maximum) values. In order to maintain an unbiased screening scenario, we did not exclude common age-related presence of well-treated diabetes, hypertension and hypothyroidism. *As per age, gender and ethnicity-based recommendations of the Bindex® software, we considered BMD scores of > 0.876 g/cm2, 0.803–0.876 g/cm2 and < 0.803 g/cm2 as normal, at risk (/osteopenia) and at high-risk of osteoporosis, respectively.

The authors declare no competing interests.

SupplementarymaterialsSukuletal.pdf
figure S1 and figure S2

Download PDF

Version 1

posted

You are reading this latest preprint version

Exhaled breath metabolites reveal postmenopausal gut-bone cross-talk and non-invasive markers for osteoporosis

Status:

Version 1

Abstract

Figures

One-Sentence Summary

Introduction

Results

Discussion

Materials And Method

Declarations

References

Tables

Additional Declarations

Supplementary Files

Status:

Version 1