The　prognostic　significance　of　the　C-reactive　protein-to-albumin　ratio　(CAR)　in　patients　undergoing　noncardiac　surgery　has　not　been　conclusively　determined.　This　study　aimed　to　investigate　the　correlation　and　predictive　capacity　of　the　CAR　for　major　perioperative　adverse　cardiovascular　events　(MACEs)　in　noncardiac　surgical　patients.　Individuals　who　underwent　noncardiac　surgical　procedures　were　identified　within　the　perioperative　medicine　database　(INSPIRE　1.1)　and　stratified　into　tertiles　on　the　basis　of　the　CAR.　The　associations　between　the　CAR　and　the　risk　of　MACEs　(all-cause　death,　perioperative　myocardial　infarction,　and　acute　heart　failure)　were　analysed　via　Cox　proportional　hazards　regression　analysis,　and　restricted　cubic　splines　were　subsequently　used　to　examine　the　shape　of　the　associations.　Additionally,　the　CAR＇s　incremental　predictive　value　for　MACEs　was　evaluated　using　the　C　statistic,　continuous　net　reclassification　improvement　(NRI),　and　integrated　discrimination　improvement　(IDI),　in　order　to　quantify　the　improvement　in　predictive　performance　achieved　by　adding　the　CAR　to　the　Revised　Cardiac　Risk　Index　(RCRI).　The　feature　significance　and　predictive　models　were　developed　via　the　Boruta　algorithm　and　machine　learning　techniques.　A　total　of　7,366　patients　were　included.　There　was　a　significant　difference　in　the　perioperative　MACE　risk　among　the　three　groups　(tertiles　1　to　3:　0.24%　vs.　0.69%　vs.　1.63%;　log-rank　P　＜　0.001).　Compared　with　patients　in　the　lowest　tertile,　those　with　the　highest　CAR　had　a　significantly　increased　risk　of　MACEs　(hazard　ratio　[HR]　4.43,　95%　confidence　interval　[CI]　1.53-12.86,　P　=　0.006),　and　an　L-shaped　association　was　observed.　For　each　SD　(standard　deviation)　increase　in　the　CAR,　the　risk　of　perioperative　MACE　increased　by　1.22　times　(HR　1.22,　95%　CI　1.06-1.41).　Furthermore,　incorporating　the　CAR　into　the　Revised　Cardiac　Risk　Index　(RCRI)　or　baseline　risk　models　significantly　enhanced　the　ability　of　the　CAR　to　predict　perioperative　MACEs.　The　CatBoost　algorithm　model　showed　the　best　performance　(AUC　=　0.865).　The　CAR　was　associated　with　an　increased　risk　of　perioperative　MACEs　in　patients　undergoing　noncardiac　surgery,　indicating　its　potential　as　a　valuable　and　reliable　prognostic　tool　for　evaluating　the　risk　of　perioperative　MACEs.