More Than Half of Readmissions Within 30 Days of Acute MI Unrelated to Index Event

Download this article's Factoid in PDF (& PPT for Gold Subscribers)

Comorbidities and complications from angiography, revascularization or reperfusion account for the majority of hospital readmissions within 30 days of acute myocardial infarction (MI), according to data from a community registry published in the July 3, 2012, issue of the Annals of Internal Medicine. The findings suggest that many rehospitalizations could be prevented by comprehensive management strategies.

Researchers led by Véronique L. Roger, MD, MPH, of the Mayo Clinic (Rochester, MN), studied 3,010 patients hospitalized for first-time MI in Olmsted County, MN, from 1987 to 2010. Using patient-level information from medical records detailing all sources of care for area residents, the investigators used the population-based registry to compare readmissions over time and as related to the index hospitalization. In all, 9.4% of the cohort underwent CABG, 51.2% had PCI, and 76.0% angiography.

Wide Variation in Readmission Predictors

Over the 23-year period, the 30-day readmission rate stayed consistent, averaging 18.6%. The rehospitalization rates by period were:

  • 1987 to 1992: 23.2%
  • 1993 to 1998: 21.9%
  • 1999 to 2004: 22.1%
  • 2005 to 2010: 18.9%

Length of stay for the first rehospitalization ranged from 0 to 64 days (median 3 days) and was similar whether readmission was related or unrelated to the index MI or its treatment.

The most common reasons for readmission were ischemic heart disease (International Classification of Diseases [ICD]-9 codes 414 and 411), respiratory or chest symptoms (ICD-9 code 786), heart failure (ICD-9 code 428) and acute MI (ICD-9 code 410). In a substantial proportion of cases, there was no direct link to the index event or related treatment; for example, the connection was unclear for more than three-quarters of patients readmitted due to respiratory and other chest symptoms (table 1).

Table 1. Cause of 30-Day Readmission and Relation to Index MI


% of Readmissions




Ischemic Heart Diseasea




Other Chest Symptoms





Heart Failure





Acute MI





Ischemic Heart Diseaseb




a ICD-9 code 414.
b ICD-9 code 411.

Overall, 42.6% of rehospitalizations after MI were related to the initial event or its treatment, whereas 30.2% were unrelated and 27.2% had an unclear relationship. Rehospitalizations related to the first admission decreased over time (from 47.5% in 1987 to 36.7% by 2010), whereas those of an unclear nature increased over the same period (from 24.8% in 1987 to 32.9 by 2010). The proportion of unrelated rehospitalizations was higher in women and patients with NSTEMI. The most common reason for rehospitalizations deemed unclear was atypical chest pain.

Diabetes, COPD, anemia, higher Killip class at presentation, longer index length of stay, and complications of angiography or reperfusion or revascularization during the index hospitalization were independently associated with increased risk for rehospitalization after MI. After adjustment for potential confounders, the risk for rehospitalization did not change over the study period.

Data Unmask High-Risk Elderly with Comorbidities

Since most of the findings presented thus far have been based on claims data, Dr. Roger told TCTMD in a telephone interview that this study brings something new to the table. “We went into this saying we have extensive medical records, we have our clinical knowledge, and we have our community cohort,” she recalled. “So we were uniquely positioned to begin to scratch the surface of the readmission issue.”

The readmission rate is unsurprising in light of what has already been documented in the literature, but what “really was of high interest to us was the fact that these early hospitalizations in fact [make-up] a large proportion of the cases not related to the heart attack,” Dr. Roger noted.

Many of the readmissions, she continued, are “simply related to who the patients are.” Because the age at which patients experience their first MI has increased over time, patients are typically elderly and laden with comorbidities. This “represents a challenge for health care providers in terms of providing the best care for these patients and this care really has to be holistic, centered around the person, and not the disease,” Dr. Roger stressed.

The paper advises that prevention of complications and close follow-up for patients who have them are particularly important ways to avoid rehospitalization.

‘A Step in the Right Direction’

According to Edward Hannan, PhD, of the University of Albany (Albany, NY), much of the study is similar to what has been found by others, especially in terms of readmission predictors. In a telephone interview with TCTMD, he said that the obvious difference in time period “shows interestingly, [but] not necessarily surprisingly, how much things have changed over the years.”

Having called for more research that takes into account specifically why readmissions occur, Jeffrey W. Moses, MD, of Columbia University Medical Center/Weill Cornell Medical Center (New York, NY), praised the study but revealed some limitations. First, since the study period reaches across several decades with a “total change of MI care and definition in that time, it’s really hard to get your arms around it,” he told TCTMD in a telephone interview.

Dr. Roger pointed out that while care has improved via more targeted treatments and powerful drugs, this shift would be expected to result in declining readmission rates. However, that pattern was not observed.

“What they don’t pick up on,” Dr. Moses added, “is that readmission rates haven’t gone up even though length of stay has dropped dramatically. [More specifically,] the shortening of the length of stay hasn’t been at the expense of more people bouncing back into the hospital. So that’s actually a positive thing about the acute care aspect.”

Puzzled by the methodology of readmission classification, Dr. Hannan observed that some of the most common reasons for rehospitalization are “the kinds of things that happen as complications in the hospital, so I don’t know that I would call that unrelated.” Examples include including urinary tract disorders and fluid and electrolyte disorders. These complications are generally unrelated to heart problems, he acknowledged, but related to the index hospitalization. “At best I would have called them unclear,” Dr. Hannan said.

According to Dr. Moses, looking at community registry data with clear-cut case examples is a “step in the right direction” but more needs to be understood about how to effectively prevent readmissions altogether. “It’s a work in progress. This is a very primitive level of data gathering, but it’s a start,” he observed.

“Readmissions are not ‘one size fits all,’” Dr. Roger concluded. “[They] are a heterogeneous group of events. Some of them are related to the original hospital admission and therefore are legitimate concerns, but a lot of them are not. Examining patterns of care in the hospital during the initial hospitalization will not address those.”

Study Details

In-hospital survival improved over time, from 89.0% in 1987 to 1992 to 95.8% in 2005 to 2010 (P < 0.001). The mean age at MI diagnosis was 67 years; 40.5% of patients were female, and 31.2% had STEMI. The frequencies of hypertension, hyperlipidemia, diabetes, obesity, COPD, and anemia increased over time. The proportion of patients with STEMI decreased from 39.1% in 1987 to 1992 to 24.1% in 2005 to 2010 (P < 0.001).


Dunlay SM, Weston SA, Killian JM, et al. Thirty-day rehospitalizations after acute myocardial infarction: A cohort study. Ann Intern Med. 2012;157:11-18.



Related Stories:

Click here for a listing of companies that provide support to the Cardiovascular Research Foundation, owner and operator of TCTMD.

More Than Half of Readmissions Within 30 Days of Acute MI Unrelated to Index Event

Comorbidities and complications from angiography, revascularization or reperfusion account for the majority of hospital readmissions within 30 days of acute myocardial infarction (MI), according to data from a community registry published in the July 3, 2012, issue of the Annals
  • Drs. Roger, Hannan, and Moses report no relevant conflicts of interest.