• 
• 

Three Part Question

In [patients with chest pain of suspected cardiac origin] does [measurement of serum PAPP-A on admission] allow [exclusion of acute coronary syndromes]?

Clinical Scenario

A forty-five year old abusive, intoxicated recurrent attendee complains of central chest pain of 2 hours duration. His initial ECG is normal. Your gut feeling is that he does not have an acute coronary syndrome. You are reluctant to admit him for troponin estimation at 12 hours but wonder if you ought to risk discharge without further investigation. Having heard about PAPP-A, a promising cardiac marker, you wonder if the evidence is sufficient to allow clinical use in this situation.

Search Strategy

OVID Medline 1966 - 2006 September Week 2
OVID Embase 1980 - 2006 Week 36
The Cochrane Library 2006 Issue 3
Medline:
[exp Myocardial Infarction/ OR exp Angina, Unstable/ OR unstable angina.mp. OR (myocard$ adj infarct$).mp. OR (myocard$ adj ischem$).mp. OR (myocard$ adj ischaem$).mp. OR acute coronary syndrome.mp. OR MI.mp. OR AMI.mp. OR ACS.mp. OR heart attack$.mp.] AND [exp Pregancy Associated Plasma Protein-A/ OR PAPP-A.mp.]
Embase:
[exp Myocardial Infarction/ OR exp Angina, Unstable/ OR exp Heart Attack/ OR exp Coronary Thrombosis/ OR (myocard$ adj (ischaem$ OR ischem$ OR infarct$)).mp. OR MI.mp. OR heart attack$.mp. OR ACS.mp.] AND [PAPP-A.mp. OR exp Pregnancy Associated Plasma Protein A/] limit to human and English language
Cochrane:
(MeSH Pregnancy Associated Plasma Protein A) OR PAPP-A (Search All Fields)

Search Outcome

Altogether 35 papers were identified using Medline, 59 using Embase and 10 using Cochrane. Five relevant papers were identified using both Medline and Embase, while no relevant papers were identified using Cochrane.

Relevant Paper(s)

Author, date and country	Patient group	Study type (level of evidence)	Outcomes	Key results	Study Weaknesses
Laterza et al 2004 USA	346 patients presenting to the Emergency Department with symptoms suggestive of acute coronary syndrome Blood taken for PAPP-A and troponin T on admission. Follow-up for 30 days	Prospective observational cohort	Diagnosis of acute MI (old WHO definition)	No significant difference in PAPP-A between MI and non-MI patients (p=0.514)	Average time from symptom onset to enrollment was 9 hours.
			Diagnosis of acute I (new definition)	Significantly higher PAPP-A in the MI group (p<0.001)
			Adverse events at 30 days (death, MI, revascularisation)	Sensitivity of PAPP-A 66.7% (95% CI 48.2-82.0%); specificity 51.1% (45.4-56.8%), PPV 12.6%, NPV 93.6%. LR+ 1.41, LR- 0.63. For every 100 patients discharged and reassured on the basis of a negative result, three would have an adverse cardiac event.
			Correlation of PAPP-A with troponin T	Poor correlation (r=0.1195)
Heeschen et al 2005 Netherlands	547 patients with available baseline blood samples, who had been enrolled in the CAPTURE trial with acute coronary syndrome (recurrent chest pain at rest associated with ECG changes during treatment with intravenous nitroglycerin and heparin). All patients underwent coronary angiography before randomisation, which demonstrated a stenosis >=70%, and subsequently underwent angioplasty. The CAPTURE trial had randomised patients to receive either abciximab or placebo. The abciximab group was excluded from this study as abciximab has been shown to interact with troponin T and sCD40L, both of which were also measured in this study. Blood had been taken 8.7+/- 4.9 hours after the last episode of chest pain. A second group of 626 consecutive patients presenting to the Emergency Department with acute chest pain <12 hours and had no ST elevation was then recuited to prospectively validate the predictive value of PAPP-A. Blood was taken 5.1 +/- 3.4 hours after symptom onset. Patients were divided into quintiles with regard to their PAPP-A results for analysis.	Observational cohort, initially retrospective but with prospective validation	Correlation of PAPP-A with troponin T	Poor correlation (r=0.11) in the clinical trial group	Results are not reported in a clinically meaningful manner. There is insufficient data to allow the calculation of sensitivities and specificities. The use of a cut-off value for PAPP-A to aid diagnosis or exclusion of acute coronary syndrome was not prospectively validated.
			Cardiac events at follow up in clinical trial group	No significant difference between the quintiles at 24h (p=0.69); significantly increased incidence with rising PAPP-A at 72 hours (p=0.019), 30 days (p=0.008) and 6 months (p=0.004)
			Cardiac events at 30 days in the Emergency Room group	Patients with PAPP-A >12.6mIU/l had significantly more events than patients with low PAPP-A (16.9% vs. 7.9%, adjusted odds ratio 2.32, p=0.008)
Lund et al 2003 Finland	136 patients who had presented to the Emergency Department with suspected acute coronary syndrome and tested negative for troponin I during the first 24 hours. Blood was sent for PAPP-A at admission, 6-12 hours and 24 hours. Patients were followed up for 6 months	Prospective observational cohort	Adverse cardiac events within 6 months according to highest PAPP-A level during admission	PAPP-A found to be an independent predictor (adjusted risk ratio 4.6, 95% CI 1.8 - 11.8, p=0.002).	The troponin positive group were excluded only when their troponin results became available. Their inclusion in the analysis would have allowed more clinically relevant conclusions to be drawn from the data. Significantly more patients who had a raised PAPP-A had a previous MI (p=0.035) or were diabetic (p=0.027) at baseline.
			Adverse cardiac event within 6 months according to admission PAPP-A level	12 of 40 (30%) vs. 14 of 96 (14.6%) had an event. This gives a sensitivity of 54%, specificity 75%, PPV 30%, NPV 15%. For every 100 patients treated on the basis of their PAPP-A result, 10 would be inappropriately reassured.
Dominguez-Rodriguez A; Abreu-Gonzalez P; Garcia-Gonzalez M; Ferrer J; Vargas M 2004 Canada	80 consecutive patients with ST elevation myocardial infarction (STEMI) and 80 healthy controls without MI. Blood taken on admission to CCU before any medications for PAPP-A.	Prospective diagnostic cohort	Diagnosis of myocardial infarction	No significant difference in PAPP-A between the MI group and the control group	No explanation of how the healthy controls were recruited. No definition of STEMI, no mention of how myocardial infarction was definitively diagnosed.
			Correlation between PAPP-A, CK-MB and troponin I	No association (CK-MB r=0.06, p=0.78; troponin I r=0.04, p=0.82)
Bayes-Genis et al 2001 USA	17 patients with acute myocardial infarction, 20 with unstable angina, 19 with stable angina, and 13 age-matched controls without clinical or angiographic evidence of coronary atherosclerosis. All patients were scheduled to undergo coronary angiography. Consecutive patients were recruited.	Prospective cohort study	Median PAPP-A levels in controls v. stable angina	Higher in stable angina but not statistically significant. Controls 7.4mIU/l (range 3.8-10.4), stable angina 8.4mIU/l (range 4.4-22.5, p=0.07).	While the methods utilised are useful for identifying promising markers for further identification, they are suboptimal to assess diagnostic efficacy in clinical practice. Recruitment of an undifferentiated group of patients who are all subjected to the same gold standard diagnostic test would be necessary to achieve this. Specificity cannot be calculated for the same reason. Small numbers.
			Median PAPP-A in unstable angina v. controls and stable angina	Significantly higher in unstable angina, when compared to both controls (P<0.001) and stable angina (P<0.001). Unstable angina 14.9mIU/l (range 6.3-63.4).
			Median PAPP-A in myocardial infarction v. controls and stable angina	Significantly higher in myocardial infarction, compared with both controsl (P<0.001) and stable angina (P<0.001). Myocardial infarction 20.6mIU/l (range 9.2-46.6)
			Median PAPP-A in myocardial infarction v. unstable angina	No significant difference (P=0.75)
			PAPP-A levels ang extent of angiographic atherosclerosis (number of vessels with clinically significant luminal stenosis)	Significant inverse association with PAPP-A (P=0.04)
			Mean area (+/- SE) under ROC curve for diagnosis	Myocardial infarction 0.94 +/- 0.03; Unstable angina 0.88 +/- 0.05
			Diagnostic performance of PAPP-A using the optimal cut-off of 10mIU/l	Myocardial infarction: Sensitivity 94.1%. Unstable angina: Sensitivity 85%. Only 1/13 controls (8%) and 5/19 stable angina (26%) patients had PAPP-A >10mIU/l

Comment(s)

PAPP-A was first found in the serum of pregnant ladies and has been used as a marker of Down's syndrome in the first trimester (Wald et al, 1999). Its abundant expression has been demonstrated in unstable but not stable atherosclerotic plaques (Bayes-Genis et al, 2001) and raised serum PAPP-A levels are associated with complex stenoses on coronary angiography (Cosin-Sales et al, 2004). Interestingly, although Bayes-Genis et al identified significant elevations of PAPP-A in patients with both myocardial infarction and unstable angina, the levels did not correlate with the number of coronary stenoses identified at angiography. This may mean that PAPP-A helps to identify soft, inflammatory plaques that are vulnerable to rupture or erosion. These plaques are more likely to undergo outward remodelling, leading to little or no arterial stenosis, while being responsible for the majority of acute coronary syndromes (Little et al, 1988). Although the early research suggests that PAPP-A may be insufficiently sensitive for detection of myoardial infarction at presentation to the Emergency Department, it remains a very promising marker for incorporation into a multimarker strategy at patient presentation. Studies consistently demonstrate that PAPP-A may predict adverse events on clinical follow-up. Levels do not correlate with the troponins, which may mean that PAPP-A helps to identify patients with unstable coronary disease without infarction. While current diagnostic strategies enable sensitive detection of infarction, the identification of patients at risk of infarction in the near future would be of tremendous benefit. Further research into PAPP-A when incorporated into a multimarker strategy for the risk stratification of patients with suspected ACS is warranted.

Clinical Bottom Line

PAPP-A may be insufficiently sensitive for the detection of myocardial infarction at presentation to the Emergency Department. However, it is a promising marker for the detection of unstable coronary disease and may have tremendous potential as a prognostic marker, especially if incorporated into a multimarker strategy.

Level of Evidence

Level 2 - Studies considered were neither 1 or 3.

References

1. Laterza OF; Cameron SJ; Chappell D; Sokoll LJ; Green GB Evaluation of pregnancy-associated plasma protein A as a prognostic indicator in acute coronary syndrome patients Clinica Chimica Acta 2004; 348: 163-169
2. Heeschen C; Cimmeler S; Hamm CW; Fichtlschere S; Simoons ML; Zeiher AM; for the CAPTURE Study Investigators Pregnancy-associated plasma protein-A levels in patients with acute coronary syndromes Journal of the American College of Cardiology 2005; 45(2): 229-237
3. Lund J; Qin QP; Ilva T; Pettersson K; Voipio-Pulkki LM; Porela P; Pulkki K Circulating pregnancy-associated plasma protein A predicts outcome in patients with acute coronary syndrome but no troponin I elevation Circulation 2003; 108: 1924-1926
4. Dominguez-Rodriguez A; Abreu-Gonzalez P; Garcia-Gonzalez M; Ferrer J; Vargas M Circulating pregnancy-associated plasma protein A is not an early marker of acute myocardial infarction Clinical Biochemistry 2005; 38: 180-182
5. Bayes-Genis A; Conover CA; Overgaard MT; Bailey KR; Christiansen M; Holmes DR; Virmani R; Oxvig C; Schwartz RS Pregnancy-associated plasma protein A as a marker of acute coronary syndromes New England Journal of Medicine 2001; 345: 1022-1029
6. Cosin-Sales J; Christiansen M; Kaminski P; Oxvig C; Overgaard MT; Cole D; Holt DW; Kaski JC Pregnancy-associated plasma protein A and its endogenous inhibitor, the proform of eosinophil major basic protein (proMBP), are related to complex stenosis morphology in patients with stable angina... Circulation 2004; 109: 1724-1728
7. Wald NJ; Watt HC; Hackshaw AK Integrated screening for Down's syndrome based on tests performed during the first and second trimesters New England Journal of Medicine 1999; 341: 461-467