1
|
Hill D, Torop M, Masoomi A, Castaldi PJ, Silverman EK, Bodduluri S, Bhatt SP, Yun T, McLean CY, Hormozdiari F, Dy J, Cho MH, Hobbs BD. Deep Learning Utilizing Suboptimal Spirometry Data to Improve Lung Function and Mortality Prediction in the UK Biobank. MEDRXIV : THE PREPRINT SERVER FOR HEALTH SCIENCES 2023:2023.04.28.23289178. [PMID: 37162978 PMCID: PMC10168495 DOI: 10.1101/2023.04.28.23289178] [Citation(s) in RCA: 0] [Impact Index Per Article: 0] [Reference Citation Analysis] [Abstract] [Grants] [Track Full Text] [Subscribe] [Scholar Register] [Indexed: 05/11/2023]
Abstract
Background Spirometry measures lung function by selecting the best of multiple efforts meeting pre-specified quality control (QC), and reporting two key metrics: forced expiratory volume in 1 second (FEV1) and forced vital capacity (FVC). We hypothesize that discarded submaximal and QC-failing data meaningfully contribute to the prediction of airflow obstruction and all-cause mortality. Methods We evaluated volume-time spirometry data from the UK Biobank. We identified "best" spirometry efforts as those passing QC with the maximum FVC. "Discarded" efforts were either submaximal or failed QC. To create a combined representation of lung function we implemented a contrastive learning approach, Spirogram-based Contrastive Learning Framework (Spiro-CLF), which utilized all recorded volume-time curves per participant and applied different transformations (e.g. flow-volume, flow-time). In a held-out 20% testing subset we applied the Spiro-CLF representation of a participant's overall lung function to 1) binary predictions of FEV1/FVC < 0.7 and FEV1 Percent Predicted (FEV1PP) < 80%, indicative of airflow obstruction, and 2) Cox regression for all-cause mortality. Findings We included 940,705 volume-time curves from 352,684 UK Biobank participants with 2-3 spirometry efforts per individual (66.7% with 3 efforts) and at least one QC-passing spirometry effort. Of all spirometry efforts, 24.1% failed QC and 37.5% were submaximal. Spiro-CLF prediction of FEV1/FVC < 0.7 utilizing discarded spirometry efforts had an Area under the Receiver Operating Characteristics (AUROC) of 0.981 (0.863 for FEV1PP prediction). Incorporating discarded spirometry efforts in all-cause mortality prediction was associated with a concordance index (c-index) of 0.654, which exceeded the c-indices from FEV1 (0.590), FVC (0.559), or FEV1/FVC (0.599) from each participant's single best effort. Interpretation A contrastive learning model using raw spirometry curves can accurately predict lung function using submaximal and QC-failing efforts. This model also has superior prediction of all-cause mortality compared to standard lung function measurements. Funding MHC is supported by NIH R01HL137927, R01HL135142, HL147148, and HL089856.BDH is supported by NIH K08HL136928, U01 HL089856, and an Alpha-1 Foundation Research Grant.DH is supported by NIH 2T32HL007427-41EKS is supported by NIH R01 HL152728, R01 HL147148, U01 HL089856, R01 HL133135, P01 HL132825, and P01 HL114501.PJC is supported by NIH R01HL124233 and R01HL147326.SPB is supported by NIH R01HL151421 and UH3HL155806.TY, FH, and CYM are employees of Google LLC.
Collapse
Affiliation(s)
- Davin Hill
- Department of Electrical and Computer Engineering, Northeastern University, Boston, MA, USA
- Channing Division of Network Medicine, Brigham and Women’s Hospital, Boston, MA, USA
| | - Max Torop
- Department of Electrical and Computer Engineering, Northeastern University, Boston, MA, USA
| | - Aria Masoomi
- Department of Electrical and Computer Engineering, Northeastern University, Boston, MA, USA
| | - Peter J. Castaldi
- Channing Division of Network Medicine, Brigham and Women’s Hospital, Boston, MA, USA
- Division of General Medicine and Primary Care, Brigham and Women’s Hospital, Boston, MA, USA
- Harvard Medical School, Boston, MA, USA
| | - Edwin K. Silverman
- Channing Division of Network Medicine, Brigham and Women’s Hospital, Boston, MA, USA
- Harvard Medical School, Boston, MA, USA
- Division of Pulmonary and Critical Care Medicine, Brigham and Women’s Hospital, Boston, MA, USA
| | - Sandeep Bodduluri
- Division of Pulmonary, Allergy and Critical Care Medicine, University of Alabama at Birmingham, Birmingham, AL, USA
| | - Surya P. Bhatt
- Division of Pulmonary, Allergy and Critical Care Medicine, University of Alabama at Birmingham, Birmingham, AL, USA
| | | | | | | | - Jennifer Dy
- Department of Electrical and Computer Engineering, Northeastern University, Boston, MA, USA
| | - Michael H. Cho
- Channing Division of Network Medicine, Brigham and Women’s Hospital, Boston, MA, USA
- Harvard Medical School, Boston, MA, USA
- Division of Pulmonary and Critical Care Medicine, Brigham and Women’s Hospital, Boston, MA, USA
| | - Brian D. Hobbs
- Channing Division of Network Medicine, Brigham and Women’s Hospital, Boston, MA, USA
- Harvard Medical School, Boston, MA, USA
- Division of Pulmonary and Critical Care Medicine, Brigham and Women’s Hospital, Boston, MA, USA
| |
Collapse
|
2
|
|
3
|
Lechtzin N, Cudkowicz ME, de Carvalho M, Genge A, Hardiman O, Mitsumoto H, Mora JS, Shefner J, Van den Berg LH, Andrews JA. Respiratory measures in amyotrophic lateral sclerosis. Amyotroph Lateral Scler Frontotemporal Degener 2018; 19:321-330. [PMID: 29566571 DOI: 10.1080/21678421.2018.1452945] [Citation(s) in RCA: 35] [Impact Index Per Article: 5.8] [Reference Citation Analysis] [Abstract] [Key Words] [Track Full Text] [Journal Information] [Subscribe] [Scholar Register] [Indexed: 12/11/2022]
Abstract
OBJECTIVE Amyotrophic lateral sclerosis (ALS) is a progressive neuromuscular disease that causes skeletal muscle weakness, including muscles involved with respiration. Death often results from respiratory failure within 3-5 years. Monitoring respiratory status is therefore critical to ALS management, as respiratory/pulmonary function tests (PFTs) are used to make decisions including when to initiate noninvasive ventilation. Understanding the different respiratory and PFTs as they relate to disease progression and survival may help determine which tests are most suitable. METHODS This review describes the tests used to assess respiratory muscle and pulmonary function in patients with ALS and the correlations between different respiratory measures and clinical outcomes measures. RESULTS The most commonly used measurement, forced vital capacity (VC), has been shown to correlate with clinical milestones including survival, but also requires good motor coordination and facial strength to form a tight seal around a mouthpiece. Other tests such as slow VC, sniff inspiratory pressure, or transdiaphragmatic pressure with magnetic stimulation are also associated with distinct advantages and disadvantages. CONCLUSIONS Therefore, how and when to use different tests remains unclear. Understanding how each test relates to disease progression and survival may help determine which is best suited for specific clinical decisions.
Collapse
Affiliation(s)
- Noah Lechtzin
- a Johns Hopkins University School of Medicine , Baltimore , MD , USA
| | | | - Mamede de Carvalho
- c Faculty of Medicine, IMM, University of Lisbon , Department of Neurosciences-CHLN , Lisbon , Portugal
| | - Angela Genge
- d Montreal Neurological Institute , Montreal , Canada
| | - Orla Hardiman
- e Trinity Biomedical Sciences Institute, Trinity College , Dublin , Ireland
| | - Hiroshi Mitsumoto
- f Eleanor and Lou Gehrig ALS Center, The Neurological Institute, Columbia University , New York , NY , USA
| | - Jesus S Mora
- g Unidad de ELA-Hospital Universitario La Paz-Hospital Carlos III , Madrid , Spain
| | - Jeremy Shefner
- h Department of Neurology , Barrow Neurological Institute , Phoenix , AZ , USA
| | - Leonard H Van den Berg
- i Department of Neurology , Brain Centre Rudolf Magnus, University Medical Centre Utrecht , Utrecht , The Netherlands
| | - Jinsy A Andrews
- f Eleanor and Lou Gehrig ALS Center, The Neurological Institute, Columbia University , New York , NY , USA
| |
Collapse
|
4
|
Holmes SJ, Allen SC, Roberts HC. Relationship between lung function and grip strength in older hospitalized patients: a pilot study. Int J Chron Obstruct Pulmon Dis 2017; 12:1207-1212. [PMID: 28458532 PMCID: PMC5402889 DOI: 10.2147/copd.s120721] [Citation(s) in RCA: 8] [Impact Index Per Article: 1.1] [Reference Citation Analysis] [Abstract] [Key Words] [MESH Headings] [Track Full Text] [Download PDF] [Figures] [Journal Information] [Subscribe] [Scholar Register] [Indexed: 12/18/2022] Open
Abstract
OBJECTIVE Older people with reduced respiratory muscle strength may be misclassified as having COPD on the basis of spirometric results. We aimed to evaluate the relationship between lung function and grip strength in older hospitalized patients without known airways disease. METHODS Patients in acute medical wards were recruited who were aged ≥70 years; no history, symptoms, or signs of respiratory disease; Mini Mental State Examination ≥24; willing and able to consent to participate; and able to perform hand grip and forced spirometry. Data including lung function (forced expiratory volume in 1 second [FEV1], forced vital capacity [FVC], FEV1/FVC, peak expiratory flow rate [PEFR], and slow vital capacity [SVC]), grip strength, age, weight, and height were recorded. Data were analyzed using descriptive statistics and linear regression unadjusted and adjusted (for age, height, and weight). RESULTS A total of 50 patients (20 men) were recruited. Stronger grip strength in men was significantly associated with greater FEV1, but this was attenuated by adjustment for age, height, and weight. Significant positive associations were found in women between grip strength and both PEFR and SVC, both of which remained robust to adjustment. CONCLUSION The association between grip strength and PEFR and SVC may reflect stronger patients generating higher intrathoracic pressure at the start of spirometry and pushing harder against thoracic cage recoil at end-expiration. Conversely, patients with weaker grip strength had lower PEFR and SVC. These patients may be misclassified as having COPD on the basis of spirometric results.
Collapse
Affiliation(s)
- Sarah J Holmes
- Medicine and Elderly Care, Hampshire Hospitals NHS Foundation Trust, Winchester
| | - Stephen C Allen
- Medicine and Geriatrics, The Royal Bournemouth Hospital and Christchurch Hospitals NHS Foundation Trust, Bournemouth
- Centre of Postgraduate Medical Research and Education, Bournemouth University, Poole
| | - Helen C Roberts
- Academic Geriatric Medicine, University of Southampton
- University Hospital Southampton NHS Foundation Trust, Southampton, UK
| |
Collapse
|
5
|
Rufino R, Costa C, Lopes A, Maiworm A, Maynard K, Silva L, Dias R. Spirometry reference values in the Brazilian population. Braz J Med Biol Res 2017; 50:e5700. [PMID: 28273211 PMCID: PMC5378453 DOI: 10.1590/1414-431x20175700] [Citation(s) in RCA: 11] [Impact Index Per Article: 1.6] [Reference Citation Analysis] [Abstract] [Key Words] [MESH Headings] [Track Full Text] [Download PDF] [Figures] [Journal Information] [Subscribe] [Scholar Register] [Received: 08/12/2016] [Accepted: 01/24/2017] [Indexed: 11/29/2022] Open
Abstract
The aim of the present study was to provide new spirometry reference equations in a sample of the Brazilian population for the following parameters: forced vital capacity (FVC), forced expiratory volume in 1 second (FEV1), FEV1/FVC ratio, peak of expiratory flow (PEF), forced expiratory flow at 50% (FEF50%), 75% average vital capacity (FEF25-75%), and average forced expiratory flow time (FEFT). This was a prospective study using results from chest radiographs, electrocardiograms, and questionnaires to investigate the participants' respiratory symptoms, sedentarism, and comorbidities (Charlson comorbidity index). From December 2010 to July 2014, individuals were randomly selected from various locations in the state of Rio de Janeiro. All individuals were examined by a single technician in the morning at the laboratory, and performed the spirometry with the same spirometer. Spirometry values were tabulated for the creation of three equation models: linear regression, logarithmic regression, and logarithms through a method that incorporates the lambda, median, and coefficient of variation (LMS method). Initially, 7003 individuals from both genders were contacted, and 454 were recruited. The data from the new equations were compared with one Brazilian and eight international equations, resulting in a high correlation (r>0.9). The values derived from the LMS method and linear regression were very similar (P>0.5), and both could be used to acquire the reference values for Brazilian spirometry. Data derived from the equations of this study were different from the current Brazilian equation, which could be justified by the different method used.
Collapse
Affiliation(s)
- R. Rufino
- Departamento de Doenças do Tórax, Faculdade de Ciências Médicas, Universidade do Estado do Rio de Janeiro, Rio de Janeiro, RJ, Brasil
| | - C.H. Costa
- Departamento de Doenças do Tórax, Faculdade de Ciências Médicas, Universidade do Estado do Rio de Janeiro, Rio de Janeiro, RJ, Brasil
| | - A.J. Lopes
- Departamento de Doenças do Tórax, Faculdade de Ciências Médicas, Universidade do Estado do Rio de Janeiro, Rio de Janeiro, RJ, Brasil
| | - A.I. Maiworm
- Departamento de Doenças do Tórax, Faculdade de Ciências Médicas, Universidade do Estado do Rio de Janeiro, Rio de Janeiro, RJ, Brasil
| | - K. Maynard
- Departamento de Doenças do Tórax, Faculdade de Ciências Médicas, Universidade do Estado do Rio de Janeiro, Rio de Janeiro, RJ, Brasil
| | - L.M.R.A. Silva
- Faculdade de Economia, Universidade Federal do Rio de Janeiro, Rio de Janeiro, RJ, Brasil
| | - R.M. Dias
- Departamento Cardiopulmonar, Escola de Medicina e Cirurgia, Universidade Federal do Estado do Rio de Janeiro, Rio de Janeiro, RJ, Brasil
| |
Collapse
|
6
|
Turkeshi E, Zelenukha D, Vaes B, Andreeva E, Frolova E, Degryse JM. Predictors of poor-quality spirometry in two cohorts of older adults in Russia and Belgium: a cross-sectional study. NPJ Prim Care Respir Med 2015; 25:15048. [PMID: 26201999 PMCID: PMC4532161 DOI: 10.1038/npjpcrm.2015.48] [Citation(s) in RCA: 8] [Impact Index Per Article: 0.9] [Reference Citation Analysis] [Abstract] [MESH Headings] [Track Full Text] [Download PDF] [Journal Information] [Subscribe] [Scholar Register] [Received: 11/13/2014] [Revised: 04/17/2015] [Accepted: 06/05/2015] [Indexed: 01/12/2023] Open
Abstract
BACKGROUND Spirometry is an important test for the diagnosis of respiratory diseases, yet it is underused especially in older adults. Several predictors of good-quality spirometry in this age group have been reported, based mainly on in/outpatients of geriatric and/or respiratory units. AIMS This study aims to assess predictors of poor-quality spirometry in community-dwelling older adults from two primary care cohorts in Russia and Belgium. METHODS Spirograms from two population-based cohort studies in Russia (CRYSTAL) and Belgium (BELFRAIL) were assessed in accordance with the American Thoracic Society/European Respiratory Society (ATS/ERS) acceptability and repeatability criteria and grouped into good and poor quality. Multivariable analysis assessed the association of poor-quality spirometry with socio-demographics, functional dependency, physical and mental functioning and co-morbidities. RESULTS In all, 43.3% of the 522 BELFRAIL participants (84.71 ± 3.67 years old) and 57.7% of the 605 CRYSTAL participants (75.11 ± 5.97 years old) achieved all ATS/ERS acceptability and repeatability criteria. In both cohorts, those with poor-quality spirometry had lower cognitive function (mini-mental state examination (MMSE) ⩽ 24). After adjustment in multivariable analysis, MMSE ⩽ 24 had an odds ratio for poor-quality spirometry of 1.33 (95% CI = 0.78-2.28) in the BELFRAIL and 1.30 (95% CI = 0.88-1.91) in the CRYSTAL cohort. CONCLUSIONS In community-dwelling older adults, including those over 80 years old, impaired cognition measured by the MMSE may not be an independent predictor of poor-quality spirometry. Further research is needed in this area, and spirometry should be used more often in older adults in primary care.
Collapse
Affiliation(s)
- Eralda Turkeshi
- Institute of Health and Society, Université Catholique de Louvain, Brussels, Belgium
| | - Dmitry Zelenukha
- Department of Family Medicine, North-Western State Medical University, St Petersburg, Russia
| | - Bert Vaes
- Institute of Health and Society, Université Catholique de Louvain, Brussels, Belgium
- Department of Public Health and Primary Care, Katholieke Universiteit Leuven, Leuven, Belgium
| | - Elena Andreeva
- Institute of Health and Society, Université Catholique de Louvain, Brussels, Belgium
| | - Elena Frolova
- Department of Family Medicine, North-Western State Medical University, St Petersburg, Russia
| | - Jean-Marie Degryse
- Institute of Health and Society, Université Catholique de Louvain, Brussels, Belgium
- Department of Public Health and Primary Care, Katholieke Universiteit Leuven, Leuven, Belgium
| |
Collapse
|
7
|
Paganoni S, Cudkowicz M, Berry JD. Outcome measures in amyotrophic lateral sclerosis clinical trials. CLINICAL INVESTIGATION 2014; 4:605-618. [PMID: 28203356 PMCID: PMC5305182 DOI: 10.4155/cli.14.52] [Citation(s) in RCA: 39] [Impact Index Per Article: 3.9] [Reference Citation Analysis] [Abstract] [Key Words] [Grants] [Track Full Text] [Subscribe] [Scholar Register] [Indexed: 02/02/2023]
Abstract
Amyotrophic lateral sclerosis (ALS) is a neurodegenerative disease with an average survival of 3-5 years. While therapies for ALS remain limited, basic and translational ALS research has been host to numerous influential discoveries in recent years. These discoveries have led to a large pipeline of potential therapies that await testing in clinical trials. Until recently, ALS clinical trials have relied on a limited cadre of 'traditional' outcome measures, including survival and measures of function. These measures have proven useful, although imperfect, in Phase III ALS trials. However, their utility in early-phase ALS trials is limited. For these early trials, outcome measures focused on target engagement or biological pathway analysis might improve trial outcomes and better support the drug development process.
Collapse
Affiliation(s)
- Sabrina Paganoni
- Harvard Medical School, Department of Neurology, Neurological Clinical Research Institute (NCRI), Massachusetts Genera Hospital, MA, USA
- Harvard Medical School, Department of Physical Medicine & Rehabilitation, Spaulding Rehabilitation Hospital, Boston, MA, USA
| | - Merit Cudkowicz
- Harvard Medical School, Department of Neurology, Neurological Clinical Research Institute (NCRI), Massachusetts Genera Hospital, MA, USA
| | - James D Berry
- Harvard Medical School, Department of Neurology, Neurological Clinical Research Institute (NCRI), Massachusetts Genera Hospital, MA, USA
| |
Collapse
|
8
|
Abstract
Spirometry is the gold standard for making the diagnosis of COPD. It should be performed in every case of suspected COPD. Other pulmonary functions, such as lung volumes, can give insight into physiologic consequences of COPD, such as hyperinflation. Pulmonary function testing can also aid in assessing the severity of disease and in managing the disease after diagnosis is made.
Collapse
Affiliation(s)
- Kristina L Bailey
- Pulmonary, Critical Care, Sleep and Allergy Division, Department of Internal Medicine, University of Nebraska Medical Center, Omaha, NE 68198-5910, USA.
| |
Collapse
|