A prognostic model for use before elective surgery to estimate the risk of postoperative pulmonary complications (GSU-Pulmonary Score): a development and validation study in three international cohorts
- Authors: Bravo L.; Simoes J.F.; Cardoso V.R.; Adisa A.; Aguilera M.L.; Arnaud A.; Biccard B.; Calvache J.; Chernbumroong S.; Elhadi M.; Ghosh D.; Gujjuri R.; Harrison E.; Ho M.W.; Kasivisvanathan V.; Kouli O.; Lederhuber H.; Li E.; Loffler M.W.; Isik A.; Marcus H.; Martin J.; McLean K.A.; Minaya-Bravo A.; Modolo M.M.; Nepogodiev D.; Pellino G.; Picciochi M.; Pockney P.; van Ramshorst G.; Riad A.; Sayyed R.; Sund M.; Gkoutos G.; Bhangu A.A.; Glasbey J.C.; Cardoso V.; Glasbey J.; Simoes J.F.F.; Kadir B.; Omar O.; Revell E.; Bahrami-Hessari M.; Ahmed W.-U.-R.; Argus L.; Ball A.; Bhangu A.; Bywater E.P.; Blanco-Colino R.; Brar A.; Chaudhry D.; Dawson B.E.; Duran I.; Gujjuri R.R.; Jones C.S.; Harrison E.M.; Kamarajah S.K.; Keatley J.M.; Lawday S.; Mann H.; Marson E.J.; Norman L.; Ots R.; Outani O.; Santos I.; Shaw C.; Taylor E.H.; Trout I.M.; Varghese C.; Venn M.L.; Xu W.; Dajti I.; Gjata A.; Kacimi S.E.O.; Boccalatte L.; Modolo M.M.; Cox D.; Aigner F.; Kronberger I.E.; Samadov E.; Alderazi A.; Padmore G.; van Ramshorst G.; Lawani I.; Cerovac A.; Delibegovic S.; Baiocchi G.; Gomes G.M.A.; Lima Buarque I.; Gohar M.; Slavchev M.; Nwegbu C.; Agarwal A.; Ng-Kamstra J.; Olivos M.; Lou W.; Ren D.-L.; Calvache J.A.; Perez Rivera C.J.; Danic Hadzibegovic A.; Kopjar T.; Mihanovic J.; Aviles Jimenez P.M.; Gouvas N.; Klat J.; Novysedlak R.; Amisi N.; Christensen P.; El-Hussuna A.; Batista S.; Lincango-Naranjo E.; Emile S.; Arevalo Sandoval D.A.; Dhufera H.; Hailu S.; Mengesha M.G.; Kauppila J.H.; Arnaud A.P.; Demetrashvili Z.; Albertsmeier M.; Loffler M.W.; Kwesi Acquah D.; Ofori B.; Tabiri S.; Metallidis S.; Tsoulfas G.; Aguilera-Arevalo M.-L.; Recinos G.; Mersich T.; Wettstein D.; Ghosh D.; Kembuan G.; Brouki Milan P.; Khosravi M.H.; Mozafari M.; Hilmi A.; Mohan H.; Zmora O.; Gallo G.; Pata F.; Fujimoto Y.; Kuroda N.; Satoi S.; Abou Chaar M.K.; Ayasra F.; Fakhradiyev I.; Hamdun I.H.S.; Jin-Young J.; Jamal M.; Karout L.; Gulla A.; Rasoaherinomenjanahary F.; Samison L.H.; Roslani A.C.; Duran Sanchez I.I.; Gonzalez D.S.; Martinez L.; Martinez M.J.; Nayen A.; Ramos-De la Medina A.; Nunez J.; Nashidengo P.R.; Shrestha A.L.; Jonker P.; Kruijff S.; Noltes M.; Steinkamp P.; Wright D.; Abdur-Rahman L.; Ademuyiwa A.; Osinaike B.; Seyi-Olajide J.; Williams O.; Williams E.; Pejkova S.; Al Balushi Z.; Qureshi A.U.; Abo Mohsen M.; Abukhalaf S.A.; Cukier M.; Gomez-Fernandez H.; Shu Yip S.; Vasquez Ojeda X.P.; Sacdalan M.D.; Major P.; Azevedo J.; Cunha M.F.; Zarour A.; Bonci E.-A.; Negoi I.; Efetov S.; Kochetkov V.; Litvin A.; Allen Ingabire J.; Bucyibaruta G.; Faustin N.; Habumuremyi S.; Imanishimwe A.; Jean de Dieu H.; Munyaneza E.; Ncogoza I.; Alameer E.; Ndong A.; Radenkovic D.; Chew M.H.; Koh F.; Ngu J.; Panyko A.; Bele U.; Kosir J.A.; Daoud H.; Minaya Bravo A.M.; Jayarajah U.; Wickramasinghe D.; Adam Essa Adam M.E.; Rutegard M.; Adamina M.; Gialamas E.; Horisberger K.; Alshaar M.; Lohsiriwat V.; Charles S.; Leventoglu S.; Lekuya H.M.; Lule H.; Kopetskyi S.; Alsaadi H.; Alshryda S.; Alser O.; Bankhead-Kendall B.; Breen K.; Kaafarani H.; Mashbari H.; Bonilla Cal F.; Al-Naggar H.; Maimbo M.; Mazingi D.; Abbott T.; Akhbari M.; Bhanderi S.; Chakrabortee S.; Costas-Chavarri A.; Demetriades A.K.; Desai A.; Di Saverio S.; Drake T.; Edwards J.; Evans J.; Fiore M.; Ford S.; Fotopoulou C.; Fowler A.; Futaba K.; Ganly I.; Grace James H.; Griffiths E.; Hutchinson P.; Hyman G.Y.; Incorvia J.; Jain R.; Jenkinson M.; Khan T.; Knight S.R.; Kolias A.; Kudsk-Iversen S.; Kwan T.Y.; Leung E.; Mayol J.; McKay S.; Meara J.G.; Mills E.; Moug S.; Patel A.; Perinotti R.; Rice H.E.; Roberts K.; Schache A.; Shaw R.; Smart N.; Stephens M.; Stewart G.D.; Teasdale E.; Vidya R.; Wright N.; Wuraola F.; Agastra E.; Thereska D.; Lucchini S.M.; Laudani V.; Chwat C.; Pedraza Salazar I.I.; Pantoja Pachajoa D.A.; Duro A.; Calderon Arancibia J.A.; Bright T.; Hollington P.; Zhou X.; Kroon H.M.; Farfus A.; Barker J.; Watson E.; Stevens S.; Latif H.; Dawson A.C.; Chuan A.; Muralidharan V.; Wong E.; Ac
- Publication year: 2024
- Type: Articolo in rivista
- OA Link: http://hdl.handle.net/10447/644154
Abstract
Background: Pulmonary complications are the most common cause of death after surgery. This study aimed to derive and externally validate a novel prognostic model that can be used before elective surgery to estimate the risk of postoperative pulmonary complications and to support resource allocation and prioritisation during pandemic recovery. Methods: Data from an international, prospective cohort study were used to develop a novel prognostic risk model for pulmonary complications after elective surgery in adult patients (aged ≥18 years) across all operation and disease types. The primary outcome measure was postoperative pulmonary complications at 30 days after surgery, which was a composite of pneumonia, acute respiratory distress syndrome, and unexpected mechanical ventilation. Model development with candidate predictor variables was done in the GlobalSurg-CovidSurg Week dataset (global; October, 2020). Two structured machine learning techniques were explored (XGBoost and the least absolute shrinkage and selection operator [LASSO]), and the model with the best performance (GSU-Pulmonary Score) underwent internal validation using bootstrap resampling. The discrimination and calibration of the score were externally validated in two further prospective cohorts: CovidSurg-Cancer (worldwide; February to August, 2020, during the COVID-19 pandemic) and RECON (UK and Australasia; January to October, 2019, before the COVID-19 pandemic). The model was deployed as an online web application. The GlobalSurg-CovidSurg Week and CovidSurg-Cancer studies were registered with ClinicalTrials.gov, NCT04509986 and NCT04384926. Findings: Prognostic models were developed from 13 candidate predictor variables in data from 86 231 patients (1158 hospitals in 114 countries). External validation included 30 492 patients from CovidSurg-Cancer (726 hospitals in 75 countries) and 6789 from RECON (150 hospitals in three countries). The overall rates of pulmonary complications were 2·0% in derivation data, and 3·9% (CovidSurg-Cancer) and 4·7% (RECON) in the validation datasets. Penalised regression using LASSO had similar discrimination to XGBoost (area under the receiver operating curve [AUROC] 0·786, 95% CI 0·774–0·798 vs 0·785, 0·772–0·797), was more explainable, and required fewer covariables. The final GSU-Pulmonary Score included ten predictor variables and showed good discrimination and calibration upon internal validation (AUROC 0·773, 95% CI 0·751–0·795; Brier score 0·020, calibration in the large [CITL] 0·034, slope 0·954). The model performance was acceptable on external validation in CovidSurg-Cancer (AUROC 0·746, 95% CI 0·733–0·760; Brier score 0·036, CITL 0·109, slope 1·056), but with some miscalibration in RECON data (AUROC 0·716, 95% CI 0·689–0·744; Brier score 0·045, CITL 1·040, slope 1·009). Interpretation: This novel prognostic risk score uses simple predictor variables available at the time of a decision for elective surgery that can accurately stratify patients’ risk of postoperative pulmonary complications, including during SARS-CoV-2 outbreaks. It could inform surgical consent, resource allocation, and hospital-level prioritisation as elective surgery is upscaled to address global backlogs. Funding: National Institute for Health Research.