Table 2

Reported discrimination and calibration of prognostic factors or prediction models for AAD

Study ID	Dissection type	Predictor	Outcome	AUC (95% CI)	P value of Hosmer-Lemeshow test	Sensitivity (%)	Specificity (%)
Prognostic factors
Liu et al (2018)27	A	Fibrinogen	In-hospital mortality	0.686 (0.585–0.787)		71.90	60.40
Zindovic et al (2018)28	A	Preoperative lactic acid levels	In-hospital mortality	0.684		56.00	72.00
		Preoperative lactic acid levels	1 year mortality	0.673		48.00	74.00
		Postoperative lactic acid levels	In-hospital mortality	0.582
		Postoperative lactic acid levels	1 year mortality	0.498
Oz et al (2017)29	A	NLR	In-hospital mortality	0.919 (0.832–1.00)		86.00	91.00
Feng et al (2017)30	A	Serum cystatin C	Long-term mortality (followed up for 909 days)	0.772 (0.692–0.839)		78.53	69.23
		hs-CRP		0.640 (0.574–0.739)		86.72	46.51
		Cystatin C, hs-CRP		0.883 (0.826–0.935)		97.44	65.92
Li et al (2016)11	A	hs-TnT	Long-term mortality (followed up for 3.5 years)	0.719 (0.621–0.803)		70.80	76.40
		hs-CRP		0.700 (0.599–0.789)		48.90	94.30
		D-dimer		0.818 (0.724–0.891)		86.10	71.40
Karakoyun et al (2015)31	A	NLR	In-hospital mortality	0.829 (0.674–0.984)		77.00	74.00
Wen et al (2019)14	A/B	NT-proBNP	In-hospital mortality	0.799 (0.707–0.891)		55.20	95.70
		Aortic diameter		0.724 (0.607–0.841)		58.60	88.20
		NT-proBNP and aortic diameter		0.832 (0.735–0.929)		79.30	84.90
Liu et al (2018)32	A/B	BUN	In-hospital mortality	0.785 (0.662–0.909)		78.90	72.20
Bennett et al (2017)33	A	Serum lactic acid level	In-hospital mortality	0.88		85.00	77.00
Bennett et al (2017)33	A	Serum lactic acid level	1 year mortality	0.81		67.00	84.00
Lafçi et al (2014)34	A/B	NLR	In-hospital mortality	0.634 (0.516–0.753)		70.00	53.00
Wen et al (2013)13	A/B	D-dimer	In-hospital mortality	0.917 (0.85–0.96)		90.30	75.90
		CRP		0.822 (0.74–0.89)		100.00	54.20
		D-dimer + CRP		0.948 (0.89–0.98)		81.90	96.80
Guo et al (2019)10	A/B	TNC	In-hospital mortality	0.884 (0.809–0.937)		83.87	83.33
		TNC +D-dimer		0.946 (0.885–0.980)		90.30	88.46
		D-dimer		0.787 (0.698–0.859)		87.19	64.10
		CRP		0.758 (0.667–0.835)		90.32	55.13
		TNC + CRP		0.909 (0.839–0.956)		90.32	74.92
Ohlmann et al (2006)12	A/B	D-dimer	In-hospital mortality	0.650 (0.584–0.716)
Zhang et al (2016)35	A	WBC	In-hospital mortality			84.60	65.90
		SBP				65.90	69.20
		NT-proBNP				80.80	51.20
		D-dimer				84.60	70.70
Li et al (2019)36	B	PLR	In-hospital mortality	0.711 (0.580–0.840)		63.00	88.00
Zhang et al (2020)37	A	UA	In-hospital mortality	0.678 (0.579–0.777)		65.00	67.10
		D-dimer		0.689 (0.589–0.790)		44.70	88.80
		age		0.616 (0.507–0.724)		37.50	90.40
		UA, D-dimer, age		0.771
Bedel et al (2019)38	A	NLR	In-hospital mortality	0.746 (0.623–0.870)		70.60	76.80
Bedel et al (2019)38	A	PLR		0.750 (0.638–0.882)		76.50	78.10
Gong et al (2019)39	A	Postoperative TnI	30-Day mortality	0.711
		Postoperative Mb		0.699
		Preoperative CK-MB		0.694
		Postoperative CK-MB		0.678
		Preoperative Creatinine		0.668
		Preoperative Mb		0.644
		Preoperative D-Dimer		0.621
		Preoperative TnI		0.618
Prediction models
Develop a model without validation
Zhang et al (2015)40	A/B	Hypotension, syncope, ischaemic complications, renal dysfunction, type A, neutrophil percentage ≥80%, surgery	In-hospital mortality	0.650	0.160
Tolenaar et al (2014)8	B	Female, age, hypotension/shock, periaortic haematoma, aortic diameter ≥5.5 cm, mesenteric ischaemia, acute renal failure, limb ischaemia	In-hospital mortality		p=0.314
Mehta et al (2002)7	A	Age, female, abrupt onset pain, abnormal ECG, any pulse deficit, kidney failure, hypotension/shock/tamponade	In-hospital mortality	0.740	p=0.750
Ghoreishi et al (2018)41	A	Lactic acid, creatinine, liver malperfusion	Operative mortality	0.750
Centofanti et al (2006)42	A	Age, coma, acute renal failure, shock and redo operation	30-Day mortality	Only reported the expected mortality and observed mortality
Santini et al (2007)43	A	Age, cardiac tamponade, hypotension, acute myocardial ischaemia, mesenteric ischaemia, acute renal failure, neurologic injury	In-hospital mortality	0.763 (0.802–0.723)		55.60	82.90
Rampoldi et al (2007)44	A	Age >70, history of aortic valve replacement,hypotension (systolic blood pressure＜100 mm Hg) or shock at presentation,migrating chest pain, preoperative cardiac tamponade,any pulse deficit,ECG with findings of myocardial ischaemia or infarction	In-hospital mortality	0.760	p=0.230
Rampoldi et al (2007)44	A	Age >70, history of aortic valve replacement,hypotension (systolic blood pressure＜100 mm Hg) or shock at presentation,migrating chest pain, preoperative cardiac tamponade, any pulse deficit, intraoperative hypotension, right ventricle dysfunction at surgery, a necessity to perform a coronary artery bypass graft	In-hospital mortality	0.810	p=0.380
Leontyev et al (2016)45	A	Age, critical preoperative state, malperfusion syndrome, coronary artery disease	In-hospital mortality	0.767 (0.715–0.819)	p=0.60
Zhang et al (2019)46	B	Hypotension, Ischaemic complications, renal dysfunction, neutrophil percentage	In-hospital mortality			86 (risk score ≥4)	78 (risk score ≥4)
Develop a model with internal validation
Macrina et al (2010)47	A	Immediate postoperative chronic renal failure, circulatory arrest time, the type of surgery on ascending aorta plus hemi-arch, extracorporeal circulation time and the presence of Marfan habitus	Long-term mortality (564±48 days)	Support vector machines:0.821, neural networks: 0.870
Macrina et al (2009)48	A	Immediate postoperative presence of dialysis in continuous, renal complications, chronic renal failure, coded operative brain protection (anterograde better than retrograde perfusion), preoperative neurological symptoms, age, previous cardiac surgery, the length of extracorporeal circulation, the operative presence of haemopericardium and postoperative enterological complications	30-Day mortality	First centre: multiple logistic regression 0.879 (0.807–0. 932)
		Immediate postoperative presence of chronic renal failure, coded operative brain protection (anterograde better than retrograde perfusion), postoperative presence of dialysis in continuous, preoperative neurological symptoms, postoperative renal complications, the length of extracorporeal circulation, age, the operative presence of haemopericardium, preoperative presence of intubation, postoperative limb ischaemia and enterological complications and the year of surgery		Second centre: multiple logistic regression 0.857 (CI: 0.785 to 0.911)
				Second centre: neural networks 0.905 (0.838–0.951)
External validation
Ge et al (2013)49	A/B	EuroSCORE II	In-hospital mortality	0.490 (0.390–0.590)	p<0.001
Yu et al (2016)50	A	Scoring systems developed by Rampoldi et al	Operative mortality	0.62
			30-Day mortality	0.56
		Scoring systems developed by Centofanti et al	Operative mortality	0.66
			30-Day mortality	0.58
		Age	Operative mortality	0.67
Vrsalovic et al (2015)9	A	CRP	In-hospital mortality	0.790 (0.784–0.796)		83.00	80.00
		IRAD score		0.740 (0.733–0.747)
		IRAD score + CRP		0.890 (0.886–0.894)

Rampoldi et al scoring system was calculated for each patient as −3.20 + (0.68 × age >70) + (1.44 × history of aortic valve replacement) + (1.17 × hypotension or shock at presentation) + (0.88 × migrating chest pain) + (0.97 × preoperative cardiac tamponade) + (0.56 × any pulse deficit) + (0.57 × ECG with findings of myocardial ischaemia or infarction).
Centofanti et al scoring system was calculated for each patient as: −2.986 + (0.771 × shock) + (0.595 × reoperation) + (1.162 × coma) + (0.778 × acute renal failure) + (0.023 × age).
AAD, acute aortic dissection; BUN, blood urea nitrogen; CK-MB, creatine kinase MB isoenzyme; CRP, C-reactive protein; hs-CRP, high-sensitivity C-reactive protein; hs-TnT, high-sensitivity cardiac troponin T; EuroSCORE II, European System for Cardiac Operative Risk Evaluation; Mb, myoglobin; NLR, neutrophil lymphocyte ratio; NT-proBNP, N-terminal pro-brain natriuretic peptide; PLR, Platelet count to lymphocyte count ratio; IRAD score, international registry of acute aortic dissection score; TNC, Tenascin-C; UA, Uric Acid.