It has been convincingly shown that the release of troponin I or troponin T is associated with a high likelihood of adverse events in patients with unstable coronary artery disease.¹

Determination of blood troponin levels is an important diagnostic aid in identifying low-risk patients who can be discharged early and high-risk patients who must be treated aggressively.² According to the recommendations of the ESC/ACC Committee for the Redefinition of Myocardial Infarction, troponin tests can be ordered for patients with acute chest pain at admission, six to nine hours later, and again 12 to 24 hours later, if results from earlier specimens are negative and the clinical index of suspicion is high.² However, these results should be interpreted with caution. Although troponins are more sensitive and specific for myocardial injury, they can be seen in other non-ischaemic cardiac conditions (advanced heart failure and acute pericarditis) and in the setting of renal insufficiency.³

Initiating the full ACS (acute coronary syndrome) protocol in patients presenting with a positive troponin is not without its risks. Indeed, the combination of aspirin 300mg, clopidogrel 600mg or prasugrel 60mg, and heparin can cause potentially life-threatening bleeding. 

Furthermore, there are also financial repercussions of inappropriate testing as the troponin test is neither 100% specific nor inexpensive. Each troponin test processed currently costs €26; which amounted to a total  cost of €127,000 for the hospital in 2012.  

Aim

The purpose of this audit was to ascertain whether troponin tests are requested appropriately for acute admissions via the emergency department (ED) in a Dublin hospital. The need for such an audit had been raised and encouraged by a consultant cardiologist, who had noticed a rising number of inappropriate troponin requests which implemented negatively on the hospital budget.

Table 1. Criteria for appropriate troponin test request(click to enlarge)

Table 2. Data collection(click to enlarge)

Method

Data was collected prospectively and randomly from all admissions to the ED. Admissions were monitored through a combination of reading ED notes or following admission, the clinical notes in the ward and medical records.

The sampling period for the first audit cycle was           January 15, 2013 to February 15, 2013. For the second audit cycle the sampling period was May 15, 2013 to June 15, 2013.

The primary outcome assessed was whether troponin tests were requested appropriately. Criteria for appropriate requests were decided after discussion with the consultant cardiologist.⁴

The secondary outcome was whether the troponin tests were requested within an appropriate time frame, ie. at admission and at 12 hours, as recommended in the NICE guidelines.⁴

The criteria for the appropriateness of a request for a troponin test was based on NICE guidelines for requesting troponin (taking into account the clinical presentation, the time from onset of symptoms and the resting 12-lead ECG findings when interpreting troponin measurements) as well as the local hospital policy set up by the consultant cardiologist (see Table 1).⁵ Data were collected using a pro-forma shown in Table 2. 

Results

In the first cycle, a total of 55 acute admissions had one or more troponin tests carried out in the ED. The average age of patients was 67.7 years. Overall, 49% of patients were male and 51% female. Of the 55 patients, one was admitted through the outpatients department and the majority through the ED. Over the one month study period, we found 32 appropriate troponin I tests. The remaining requests were deemed inappropriate as per the criteria outlined in Table 1. This means that 23 tests (41.8%) of the troponin I tests requested for consecutive admissions over the audit’s first cycle period, were requested inappropriately. The cost of the inappropriate troponin I tests totalled €600.

Interestingly, of the 55 patients who had a troponin test carried out in the first cycle, only three patients had a positive troponin, two of whom proved not to have ACS after invasive cardiology testing – one had high troponin because of acute kidney injury secondary to sepsis; the other patient had a completely normal angiogram and further tests confirmed perimyocarditis. The third patient with a positive troponin in the context of new LBBB (left bundle branch block) and was eventually deemed not suitable for invasive tests.

The average time for the first troponin test was 9.45 hours (NICE guidelines recommend repeat troponin testing at 12 hours).

Of all patients who had a troponin test, five were started inappropriately on ACS protocol.

The cost of troponin tests in the hospital totalled €127,000 for 2012; and in 2013 up to the end of the second cycle of this audit (June 2013), the tests cost the hospital €51,000.

Following a re-audit, we found that from the total of 55 acute admissions through ED during the second cycle of data collection (May 15, 2013 to June 15, 2013), only 12 were inappropriate according to the criteria set up earlier (Table 1); the average time for requesting initial troponin was 9.1 hours; and only two patients in this cycle were started on ACS protocol unnecessarily. The three other patients who started on ACS protocol truly needed to do so.

Discussion 

Although there were still 12 inappropriate troponin I requests in the second audit cycle, educational measures adopted after completing the first cycle such as grand round meetings as well as circulation of the appropriateness criteria among ED staff resulted in a significant reduction (by almost half) in the number of inappropriate troponin tests requested in the ED during the second cycle (12 (21.8%) in the second cycle compared to 23 (41.8%)in the first cycle).

In terms of the time-frame for requesting troponin, we noticed that in both cycles the time during which troponin was requested was reasonable and more or less complied with the NICE guidelines – the average time that the troponin was requested in the first and second cycle was 9.45 and 9.1 hours respectively. In order to simplify the analysis, the above result was based on the assumption that any troponin test performed beyond 12 hours of onset of symptoms was regarded as a 12-hour troponin irrespective of the length of time over the 12 hour threshold.

We noticed in light of this audit that the ED staff are complying with the criteria outlined above for requesting a troponin test and accordingly they initiated their own audit cycle. This is as a part of a larger audit for other tests.

In conclusion, we demonstrated that adequate educational measures aimed at healthcare professionals can have a positive effect on appropriateness of test requests and that can have an impact not only in providing high-quality clinical care, but also on healthcare economics.

References

1. Willerson JT, Cohn JN. Cardiovascular Medicine
2. Meng QH, Zhu S, Booth C et al. Impact of the cardiac troponin testing algorithm on excessive and inappropriate troponin test requests. Am J Clin Pathol 2006; 126:195-199 
3. Griffin BP et al. Manual of Cardiovascular Medicine, Lippincott Williams & Wilkins
4. Doolub G, Hadley G, Dwight J. Troponin testing in the emergency setting: how good are we? J Clin Exp Cardiolog 2012, S:12
5. Local hospital policy for management of patients with acute coronary syndromes (adopted from ESC Guidelines)