1
|
Akita K, Kageyama S, Suzuki S, Ohno K, Kamakura M, Nawada R, Takanaka C, Wakabayashi Y, Kanda T, Tawarahara K, Mutoh M, Matsunaga M, Suwa S, Takeuchi Y, Sakamoto H, Saito H, Hayashi K, Wakahara N, Unno K, Ikoma T, Sato R, Iguchi K, Satoh T, Sano M, Suwa K, Naruse Y, Ohtani H, Saotome M, Maekawa Y. Machine learning-based detection of sleep-disordered breathing in hypertrophic cardiomyopathy. Heart 2024; 110:954-962. [PMID: 38589224 DOI: 10.1136/heartjnl-2023-323856] [Citation(s) in RCA: 0] [Impact Index Per Article: 0] [Reference Citation Analysis] [Abstract] [Key Words] [MESH Headings] [Track Full Text] [Journal Information] [Submit a Manuscript] [Subscribe] [Scholar Register] [Received: 12/28/2023] [Accepted: 03/26/2024] [Indexed: 04/10/2024] Open
Abstract
BACKGROUND Hypertrophic cardiomyopathy (HCM) is often concomitant with sleep-disordered breathing (SDB), which can cause adverse cardiovascular events. Although an appropriate approach to SDB prevents cardiac remodelling, detection of concomitant SDB in patients with HCM remains suboptimal. Thus, we aimed to develop a machine learning-based discriminant model for SDB in HCM. METHODS In the present multicentre study, we consecutively registered patients with HCM and performed nocturnal oximetry. The outcome was a high Oxygen Desaturation Index (ODI), defined as 3% ODI >10, which significantly correlated with the presence of moderate or severe SDB. We randomly divided the whole participants into a training set (80%) and a test set (20%). With data from the training set, we developed a random forest discriminant model for high ODI based on clinical parameters. We tested the ability of the discriminant model on the test set and compared it with a previous logistic regression model for distinguishing SDB in patients with HCM. RESULTS Among 369 patients with HCM, 228 (61.8%) had high ODI. In the test set, the area under the receiver operating characteristic curve of the discriminant model was 0.86 (95% CI 0.77 to 0.94). The sensitivity was 0.91 (95% CI 0.79 to 0.98) and specificity was 0.68 (95% CI 0.48 to 0.84). When the test set was divided into low-probability and high-probability groups, the high-probability group had a higher prevalence of high ODI than the low-probability group (82.4% vs 17.4%, OR 20.9 (95% CI 5.3 to 105.8), Fisher's exact test p<0.001). The discriminant model significantly outperformed the previous logistic regression model (DeLong test p=0.03). CONCLUSIONS Our study serves as the first to develop a machine learning-based discriminant model for the concomitance of SDB in patients with HCM. The discriminant model may facilitate cost-effective screening tests and treatments for SDB in the population with HCM.
Collapse
Affiliation(s)
- Keitaro Akita
- Division of Cardiology, Internal Medicine III, Hamamatsu University School of Medicine, Hamamatsu, Shizuoka, Japan
| | - Shigetaka Kageyama
- Department of Cardiology, Shizuoka City Shizuoka Hospital, Shizuoka, Japan
| | - Sayumi Suzuki
- Division of Cardiology, Internal Medicine III, Hamamatsu University School of Medicine, Hamamatsu, Shizuoka, Japan
| | - Kazuto Ohno
- Division of Cardiology, Internal Medicine III, Hamamatsu University School of Medicine, Hamamatsu, Shizuoka, Japan
| | - Masamitsu Kamakura
- Department of Cardiology, Shizuoka City Shizuoka Hospital, Shizuoka, Japan
| | - Ryuzo Nawada
- Department of Cardiology, Shizuoka City Shizuoka Hospital, Shizuoka, Japan
| | | | - Yasushi Wakabayashi
- Department of Cardiology, Seirei Mikatahara Hospital, Hamamatsu, Shizuoka, Japan
| | - Takahiro Kanda
- Department of Cardiology, Hamamatsu Red Cross Hospital, Hamamatsu, Shizuoka, Japan
| | - Kei Tawarahara
- Department of Cardiology, Hamamatsu Red Cross Hospital, Hamamatsu, Shizuoka, Japan
| | - Masahiro Mutoh
- Department of Cardiology, Hamamatsu Medical Center, Hamamatsu, Shizuoka, Japan
| | - Masaki Matsunaga
- Department of Cardiology, Iwata City Hospital, Iwata, Shizuoka, Japan
| | - Satoru Suwa
- Department of Cardiovascular Medicine, Juntendo University Shizuoka Hospital, Izunokuni, Shizuoka, Japan
| | - Yasuyo Takeuchi
- Department of Cardiology, Shizuoka General Hospital, Shizuoka, Japan
| | - Hiroki Sakamoto
- Department of Cardiology, Shizuoka General Hospital, Shizuoka, Japan
| | - Hideki Saito
- Department of Cardiology, Seirei Hamamatsu General Hospital, Hamamatsu, Shizuoka, Japan
| | - Kazusa Hayashi
- Department of Internal Medicine, JA Shizuoka Kohseiren Enshu Hospital, Hamamatsu, Shizuoka, Japan
| | - Nobuyuki Wakahara
- Department of Cardiology, Fujinomiya City General Hospital, Fujinomiya, Shizuoka, Japan
| | - Kyoko Unno
- Division of Cardiology, Internal Medicine III, Hamamatsu University School of Medicine, Hamamatsu, Shizuoka, Japan
| | - Takenori Ikoma
- Division of Cardiology, Internal Medicine III, Hamamatsu University School of Medicine, Hamamatsu, Shizuoka, Japan
| | - Ryota Sato
- Division of Cardiology, Internal Medicine III, Hamamatsu University School of Medicine, Hamamatsu, Shizuoka, Japan
| | - Keisuke Iguchi
- Division of Cardiology, Internal Medicine III, Hamamatsu University School of Medicine, Hamamatsu, Shizuoka, Japan
| | - Terumori Satoh
- Division of Cardiology, Internal Medicine III, Hamamatsu University School of Medicine, Hamamatsu, Shizuoka, Japan
| | - Makoto Sano
- Division of Cardiology, Internal Medicine III, Hamamatsu University School of Medicine, Hamamatsu, Shizuoka, Japan
| | - Kenichiro Suwa
- Division of Cardiology, Internal Medicine III, Hamamatsu University School of Medicine, Hamamatsu, Shizuoka, Japan
| | - Yoshihisa Naruse
- Division of Cardiology, Internal Medicine III, Hamamatsu University School of Medicine, Hamamatsu, Shizuoka, Japan
| | - Hayato Ohtani
- Division of Cardiology, Internal Medicine III, Hamamatsu University School of Medicine, Hamamatsu, Shizuoka, Japan
| | - Masao Saotome
- Division of Cardiology, Internal Medicine III, Hamamatsu University School of Medicine, Hamamatsu, Shizuoka, Japan
| | - Yuichiro Maekawa
- Division of Cardiology, Internal Medicine III, Hamamatsu University School of Medicine, Hamamatsu, Shizuoka, Japan
| |
Collapse
|
3
|
Zhou RF, Liang NP, Chen S, Zhang WC, Wang YX, Wang Y, Ji HF, Dong YF. Interactions Between Body Mass Index and Glomerular Filtration Rate Increase the Identification Ability of Obstructive Sleep Apnea in Patients with Hypertrophic Cardiomyopathy. Nat Sci Sleep 2022; 14:1699-1708. [PMID: 36176567 PMCID: PMC9514299 DOI: 10.2147/nss.s360317] [Citation(s) in RCA: 0] [Impact Index Per Article: 0] [Reference Citation Analysis] [Abstract] [Key Words] [Track Full Text] [Download PDF] [Figures] [Journal Information] [Submit a Manuscript] [Subscribe] [Scholar Register] [Received: 01/30/2022] [Accepted: 09/06/2022] [Indexed: 11/23/2022] Open
Abstract
PURPOSE Obstructive sleep apnea (OSA) is common in hypertrophic cardiomyopathy (HCM) patients and is related to worse adverse prognosis in HCM patients. However, there are no acknowledged warning characteristics to help to identify OSA in HCM patients. METHODS Seventy-one HCM patients and forty-nine hypertensive (HTN) patients as control group underwent polysomnography (PSG) examination at the Second Affiliated Hospital of Nanchang University from January 2015 to December 2019 patients were consecutively enrolled. The characteristics were analyzed and compared between HCM patients with OSA and without OSA. RESULTS A total of 37 (52%) HCM patients and 25 (51%) HTN patients were diagnosed with OSA. High body mass index (BMI) (OR = 1.228, 95% CI: 1.032,1.461, P = 0.020) and low estimated glomerular filtration rate (eGFR) (OR = 0.959, 95% CI: 0.931,0.989, P = 0.007) independently correlated with the occurrence of OSA in HCM patients, respectively. Multiplicative interaction was shown between high BMI and low eGFR on the risk of OSA in HCM patients (OR: 6.050, 95% CI: 1.598, 22.905, P = 0.008). The additive interaction analysis further suggested that 70.1% of HCM patients developed OSA due to the additive interaction between BMI and eGFR. The identification ability of OSA in HCM patients was significantly enhanced by using both BMI and eGFR (area under receiver-operating characteristic analysis curve 0.785; P = 0.000038) as compared with BMI (area under curve 0.683, P = 0.008) or eGFR (area under curve 0.700, P = 0.004), respectively. CONCLUSION High BMI or low eGFR independently related to the occurrence of OSA in HCM patients, and the multiplicative and additive interactions between BMI and eGFR increased the identification ability of OSA in HCM patients.
Collapse
Affiliation(s)
- Rui-Fei Zhou
- Department of Cardiovascular Medicine, the Second Affiliated Hospital of Nanchang University, Nanchang, Jiangxi, People's Republic of China.,Department of Cardiovascular Medicine, Yichun People's Hospital, Yichun, Jiangxi, People's Republic of China
| | - Ning-Peng Liang
- Department of Cardiovascular Medicine, the Second Affiliated Hospital of Nanchang University, Nanchang, Jiangxi, People's Republic of China
| | - Shuo Chen
- Department of Cardiovascular Medicine, the Second Affiliated Hospital of Nanchang University, Nanchang, Jiangxi, People's Republic of China
| | - Wen-Chao Zhang
- Department of Cardiovascular Medicine, the Second Affiliated Hospital of Nanchang University, Nanchang, Jiangxi, People's Republic of China
| | - Yi-Xi Wang
- Department of Cardiovascular Medicine, the Second Affiliated Hospital of Nanchang University, Nanchang, Jiangxi, People's Republic of China
| | - Yu Wang
- Department of Cardiovascular Medicine, the Second Affiliated Hospital of Nanchang University, Nanchang, Jiangxi, People's Republic of China
| | - Hui-Fang Ji
- Department of Cardiovascular Medicine, the Second Affiliated Hospital of Nanchang University, Nanchang, Jiangxi, People's Republic of China
| | - Yi-Fei Dong
- Department of Cardiovascular Medicine, the Second Affiliated Hospital of Nanchang University, Nanchang, Jiangxi, People's Republic of China.,Key Laboratory of Molecular Biology in Jiangxi Province, Nanchang, Jiangxi, People's Republic of China
| |
Collapse
|