HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS		QUICK SEARCH:   [advanced] Author: Keyword(s): Year:  Vol:  Page:

This Article Abstract Full Text (PDF) Alert me when this article is cited Alert me if a correction is posted Citation Map Services Email this article to a friend Similar articles in this journal Similar articles in PubMed Alert me to new issues of the journal Add to Personal Folders Download to citation manager Author home page(s): Clinton E. Baisden Permission Requests Citing Articles Citing Articles via HighWire Citing Articles via Google Scholar Google Scholar Articles by Baisden, C. E. Articles by Riggs, M. W. Search for Related Content PubMed PubMed Citation Articles by Baisden, C. E. Articles by Riggs, M. W. Related Collections Coronary disease

Ann Thorac Surg 2003;75:68-73
© 2003 The Society of Thoracic Surgeons

---

Original article: cardiovascular

Readmission and mortality in patients discharged the day after off-pump coronary bypass surgery

^a Division of Cardiothoracic Surgery, Temple, TX, USA
^b Department of Biostatistics, Texas A & M University System Health Science Center, Scott and White Memorial Hospital, Temple, Texas, USA

* Address reprint requests to Dr Baisden, Division of Cardiothoracic Surgery, Scoot and White Memorial Hospital, 2401 South 31st St, Temple, TX 76508, USA
e-mail: cbaisden{at}swmail.sw.org

Presented at the Forty-eighth Annual Meeting of the Southern Thoracic Surgical Association, San Antonio, TX, Nov 8–10, 2001.

	Abstract

Top Abstract Introduction Material and methods Results Comment Discussion References

 
BACKGROUND: The fate of patients discharged the day following off-pump coronary bypass (OPCAB) has not previously been reported. We studied the mortality and readmissions of a consecutive series of patients discharged after OPCAB, and compared the outcomes of those discharged the day following surgery to the rest of the group.

METHODS: All patients having OPCAB through median sternotomy during the calendar year 2000 by a single surgeon were retrospectively reviewed. Demograghics, intraoperative variables and postoperative complications, readmissions and mortality were recorded. Factors were analyzed to determine associations with time of discharge and readmission.

RESULTS: One hundred fifteen patients had isolated OPCAB averaging 3.1 grafts. Two patients (1.8%) died before discharge. Sixty-three of 113 patients (55.8%) were discharged on day 1 and 8 (12.7%) required readmission compared to 13 of 50 (26%) discharged later. Diabetes (p = 0.04) and renal failure (p = 0.01) exhibited univariate association with day 1 discharge while multivariate analysis added infarction. The combination of previous bypass, obesity, acute myocardial infarction, and hypertension was associated with readmission in the entire OPCAB group but not in day 1 discharged patients.

CONCLUSIONS: The readmission rate for the entire group (18.6%) was high but lower in day 1 discharge patients (12.7%). Day 1 discharge (55.8%) was unusual in patients with diabetes, renal failure, or recent infarction. Previous bypass, obesity, acute myocardial infarction, and hypertension were associated with readmission for the entire group only. Day 1 discharged patients had no deaths or serious consequences, and there were no readmissions in more than 87%.

	Introduction

Top Abstract Introduction Material and methods Results Comment Discussion References

 
Interest in short hospital stay after coronary bypass surgery surged in the early 1990s due to reductions in reimbursement for various operative procedures. Clinical pathways became popular and discharge was based on clinical events rather than the postoperative day.

Off pump coronary bypass surgery (OPCAB) was developed as an attempt to avoid the cost of the heart-lung machine and the subsystem dysfunction caused by cardiopulmonary bypass (CPB). It was anticipated OPCAB would be cheaper, allow faster recovery, and earlier hospital discharge.

One of our surgeons (JRB) developed an interest in OPCAB and took the lead in developing our experience. His coronary bypass practice evolved over 2.5 years to essentially 100% OPCAB surgery. During this time he noted some of his patients appeared ready for discharge the day after OPCAB. Slowly he allowed some of these patients to go home on postoperative day one (POD1). By the second year of this practice he was allowing about one-half of his OPCABs to be discharged on POD1.

Little, if anything, is written about the fate of patients discharged on POD1 after OPCAB. We conducted a retrospective review of these patients to determine the percent of first day discharges, their readmissions, and mortality. We analyzed several preoperative factors to determine if their presence or absence was associated with POD1 discharge or the need for readmission.

	Material and methods

Top Abstract Introduction Material and methods Results Comment Discussion References

 
During the calendar year from January 1 through December 31, 2000, 115 consecutive patients underwent isolated CABG without cardiopulmonary bypass at Scott and White Hospital by a single surgeon (JRB). These patients constitute the initial study group. Scott and White Hospital is a 400-bed facility associated with a primary multi-specialty clinic of 520 physicians and nineteen satellite clinics. Essentially all patients having heart surgery at Scott and White also have their follow-up and any subsequent care including rehospitalizations within this mostly closed system. Hospital records including admission notes, consultations, operation reports, pathology reports, laboratory data, radiology, echocardiography, and nuclear scan reports, as well as discharge summaries are all contained in an electronic medical record available for review within this system. Following institution review board approval the charts and electronic medical records of these patients were retrospectively reviewed. The preoperative comorbidities studied included age, gender, previous bypass surgery, left main more than 50%, ejection fraction less than 40%, intraaortic balloon pump (IABP) placed preoperatively, congestive heart failure, obesity, history of hypertension, diabetes, renal insufficiency (creatinine > 2.0 mg/dl), previous stroke, chronic obstructive pulmonary disease, peripheral vascular disease, previous myocardial infarction, elective, urgent or emergent status.

The operation was performed in a standard fashion by a single surgeon and one of sixteen staff anesthesiologists aided by either an anesthesia resident in training or a nurse anesthetist. All operations were performed through a full median sternotomy and the Genzyme stabilizing platform (Genzyme Corporation, Tucker, GA) was used exclusively after a rolltrack retractor was used to take down the internal mammary artery under direct vision for grafting. Minimally invasive technique was used to harvest the greater saphenous vein usually employing the Saphlite retractor (Genzyme Corporation) to aid visualization, dissection and hemostasis.

Half- to full-dose Heparin (Wyeth-Ayerst Pharmaceuticals, Philadelphia, PA) was administered before dividing the internal mammary artery then proximal vein graft anastomosis were constructed on the aorta using a partial occlusion clamp. Distal anastomoses were then performed usually to collateralized vessels first. Silicone elastomer vessel loops were used proximally and distally and any further bleeding was controlled with a sterile humidified carbon dioxide blower. Protamine sulfate was generally administered at the end of the procedure to reverse the Heparin.

Intraoperative and postoperative variables reviewed included number of distal grafts, use of the internal mammary artery, placement of the IABP during surgery, ventilation longer than 24 hours after surgery, postoperative stroke, myocardial infarction, renal failure, sternal wound infection, leg wound infection, pneumonia, sepsis, reexploration for bleeding, transfusion requirements, atrial fibrillation, and postoperative length of stay, readmission within 30 days of surgery, and death (within 30 days or during same hospitalization).

The decision to discharge the patient was made based on clinical and social grounds. Clinical requirements included stable normal heart rate and sinus rhythm, stable normal blood pressure off intravenous drips, normal respirations with clear lungs and transcutaneous oxygen saturation more than 92% on room air, ability to ambulate freely around the hospital without assistance, lightheadedness, marked shortness of breath; demonstrated ability to take hospital food and oral medications easily and regularly without nausea or emesis, no difficulty with urination, mental confusion, or anxiety. Socially the patient and his family had to express a desire and motivation to be discharged and a willingness to take responsibility in helping and encouraging the patient with medications, lung exercises, daily progressive ambulation, personal hygene, and general observation. Before discharge the surgeon (JRB), a nurse clinician, and the patient’s floor nurse individually met with each patient and their family to discuss what to expect in the postoperative period, review medications, activity level, and alert to possible complications that would require a return to the hospital. A printed sheet with this specific information for the individual patient as well as a packet of preprinted general information was given to each patient/family.

The charts and electronic medical records of all discharged patients were reviewed 30 days after their date of surgery. Post discharge hospital readmissions and the reason for readmission was recorded. Patients whose status was unclear from the records were contacted by telephone by the data manager to inquire about the patient. Any interim readmissions or deaths were noted and details were obtained. The electronic medical records of all patients were again reviewed and updated 9 and 18 months after December 2000 for accuracy and completeness.

Statistical analysis consisted of calculating the frequency, mean, and standard deviation for independent and dependent variables. Odds ratios, with 95% confidence intervals, were computed for analysis of preoperative factors associated with day of discharge or need for readmission. Multiple logistic regression using stepwise variable entry was used for multivariate analyses of preoperative factors associated with day of discharge or need for readmission.

	Results

Top Abstract Introduction Material and methods Results Comment Discussion References

 
There were 115 patients in the study group. Their mean age was 66 ± 11 years and ranged from 37 to 84 years old. Very few had previous heart surgery (7%), low ejection fraction (12%), congestive heart failure (17%), intraaortic balloon pump preoperatively (6%), chronic obstructive pulmonary disease (4%), peripheral vascular disease (12%), previous stroke (8%), renal failure (8%), urgent (12%), or emergency surgery (3%). A modest number of the patients had left main disease (26%), obesity (37%), diabetes (33%), or an acute infarction within 7 days (24%). Most of the study group was comprised of men (70%) with hypertension (77%) who had an elective (85%), first coronary operation (93%) (Table 1).

There were two 30-day/same hospitalization deaths (1.8%). One patient died of a massive stroke and one from respiratory failure.

There were no vein donor site or sternal wound infections, reoperations for bleeding, or postoperative need for the intraaortic balloon pump (Table 3).

Table 4 illustrates the number and (percent) of patients discharged on each POD and the incidence of readmission. Over half of the patients (63/113, 55.8%) were discharged on POD1 and 8/63 (12.7%) required readmission. Comparing readmission rates of those discharged on POD1 to all other days combined (POD 2 to 45) reveals POD1 to be lower (12.7% vs 26%).

Another patient (#3) was readmitted after 2 days. He suffered syncope and hypotension from a medication error. He accidentally received double doses of diuretics and ß-blockers in addition to his usual antihypertensive medications. He was readmitted, medications were held for 24 hours, restarted at the proper dose, and he was discharged home on the second readmission day. Patients #4 and #8 developed atrial fibrillation and breathing difficulties and required readmission 4 and 19 days after surgery. They were both treated successfully and discharged after a 4-day hospitalization. Patient #5 developed an upper gastrointestinal hemorrhage on POD 8 and required blood transfusions and 2 days admission. Patient #6 was admitted for severe epigastric distress on the POD 11. He was ruled out for myocardial ischemia or necrosis, treated successfully for gastroesophageal reflux, and discharged after 3 days. The next patient (#7) was admitted and treated for pneumonia 17 days after OPCAB. The readmission required 2 hospital days. Finally patient #8 was admitted on POD 19 for atrial fibrillation and congestive heart failure, treated, and released 4 days later.

An analysis of preoperative factors (Table 1) and day of discharge can be seen in Table 6. Individual factors associated with discharge after POD 1 were identified using univariate odds ratios. This study demonstrated the presence of diabetes (p = 0.04) and renal failure (p = 0.01) decreased the odds of discharge on POD1, and there was a strong trend for myocardial infarction within 7 days before surgery (p = 0.07). Multiple logistic regression was applied using stepwise variable entry to identify the combination of variables that best explained the occurrence of POD 1 discharge. The absence of diabetes, left main more than 50%, acute myocardial infarction within 7 days before surgery, and renal failure were significantly associated.

	Comment

Top Abstract Introduction Material and methods Results Comment Discussion References

 
This study was undertaken to discover the fate of a group of patients who were discharged home from the hospital the day following OPCAB. We wanted to know if any of the patients died after discharge and the percentage and reason of readmission. We felt our hospital system consisting of a central hospital and 19 regional clinics, our own health plan insurance, and our electronic medical record afforded us an opportunity to gather data that some other systems may not have. Most importantly, we had a single surgeon who performed all his coronary bypass surgeries off pump and did not hesitate to allow his patients to be discharged on the first postoperative day if he felt it was safe and they wanted to go home.

Short hospitalization after coronary bypass surgery has evolved over the past 10 to 15 years. In the early to mid 1980s the average length of stay after coronary bypass operations was 10 to 11 days [1, 2]. The main reason for postoperative hospitalization then was to treat noncardiac subsystem dysfunction and a tradition of in-hospital observation after surgery. In the 1990s rising costs and decreasing revenues made in-hospital observation of uneventful recovery unaffordable. Additionally, a method of prevention or rapid correction of postoperative noncardiac dysfunction was described in 1990 by Krohn and associates. It included pre- and postoperative steroids, early extubation, ambulation, exercise testing, etc [3]. They decreased the median length of stay after operation to 4 days in patients who had undergone coronary artery bypass, aortic, and mitral valve operations. An interesting additional finding was that the patients with the shortest initial hospitalization had the fewest rehospitalizations when followed up to 2 years.

Further decreases in length of stay were made when three-day discharge was planned and reported by Ott and colleagues [4] in a series of 104 consecutive coronary bypass patients using CPB in 1997. They used a "rapid recovery" protocol that emphasized short CPB time, short acting anesthetic agents, early extubation, steroid and thyroid hormone administration, and aggressive diuresis. Applying this protocol they were able to identify a group of patients (29%) who were successfully discharged within 3 days. This group was younger and had fewer comorbid conditions than the rest of the patients. Their readmission rate was only 3.3%. They concluded there was a group of patients with minimal comorbid conditions undergoing on pump coronary bypass surgery that could safely be discharged home within 3 days using their protocol.

Since OPCAB theoretically causes less subsystem dysfunction than on-pump surgery, it was hoped OPCAB would allow even earlier discharge. Puskas and associates at Emory compared a group of 200 consecutive OPCAB patients with a computer-matched control group seen in 1997 through 1999 [5]. OPCAB reduced the postoperative hospital stay from 5.7 ± 5.3 days to 3.9 ± 2.6 days (p < 0.001) in their patients. They noted fewer complications in the OPCAB group, which they felt accounted for the earlier discharges. They did not indicate if any of their patients were discharged on POD1 but it seems possible given the overall short average length of stay and standard deviation.

The current study reveals a subset of OPCAB patients who were safely discharged on POD1. Over half of the entire group (56%) was discharged on POD 1 and there were no deaths in them or the rest of the patients at 30 days after surgery. The readmission rate for the entire group (18.6%) is admittedly high. It was higher than that reported after isolated coronary bypass surgery in the Society of Thoracic Surgeons (STS) National Database for the past three years by 10% to 12% [6].

It does not appear that any complications were caused by POD1 discharge. However, three of the readmissions (#1, 2, and 3) may have been avoided if the patients were not discharged on POD1 (Table 5). In these cases readmission would have been traded for longer length of stay. Even so, the readmission rate of all patients discharged on POD 1 was lower than that of the remaining patients discharged after POD1.

There seems to be a subset of patients that better tolerates major surgery; our data and that of Krohn [3] would support that contention since they observed less subsequent hospitalizations in their earlier discharged patients. More than 87% of our POD1 discharges were successful and uneventful. That is comparable to the 91.7% to 93.3% nonreadmission rates for the past three years found in the STS database for patients having isolated coronary bypass surgery. These STS patients had mean postprocedure lengths of stay of 6.9 days compared to 1 day in our POD1 study patients [6].

A search for factors that would predict successful POD1 discharge yielded results that were not surprising. Obviously the initial study group would have to be considered a low risk group of patients since the majority were men, elective, first time isolated CABG patients with good ejection fractions (Table 1). The analysis showed the slight increase in mortality risk generally considered associated with diabetes mellitus, left main disease, recent myocardial infarction, and renal failure also precluded POD 1 discharge. The reason for this is uncertain but probably these factors are just markers for preoperative subsystem dysfunction that is amplified following the stress of surgery thus preempting the previously stated objective requirements for POD 1 dismissal. Why other factors known to be risk factors for mortality such as COPD, congestive heart failure, etc were not statistically indicated is probably only because of the small numbers in our data set. This is certainly a major weakness of the study. It could be argued that the lower the risk of mortality for an individual patient, the higher the probability of early discharge. It seems this question could be best answered by applying this analysis to a much larger data set such as the STS national database.

We found readmission was only associated with some preoperative factors (previous CABG, recent myocardial infarction, hypertension) when considering all readmissions together but the statistical significance was lost when considering only the POD 1 readmissions (Table 7). Since there were only 8 POD 1 readmissions compared to 21 total readmissions it is probably again just a function of analyzing too few cases.

In summary, this study reports that it was possible for a single surgeon to identify a favorable group of OPCAB patients for discharge the day following surgery. Over half the patients (56%) were in this group. Follow-up revealed the readmission rate was higher than that previously reported in the STS database. Still there were no deaths or serious consequences of POD 1 discharge and more than 87% of the patients were safely and successfully discharged. Further analysis suggested the lowest mortality risk patients had the highest chance of early discharge without need for readmission but this conclusion was limited by the relatively small number of patients in the data set.

	Discussion

Top Abstract Introduction Material and methods Results Comment Discussion References

 
DR FREDERICK L. GROVER (Denver, CO): Thanks, Clint, for sharing this and your and Randy’s experience with us. I think it raises the bar even higher and demonstrates how one can improve not only efficiency, but do it without sacrifice in the quality of care. I think, as you mentioned, your particular group of patients that you operate on do tend to be low risk, highly selected patients. I looked up, compared to the Society of Thoracic Surgeons and the VA data sets, which probably reflect a fairly broad spectrum of practice; and there is a considerably higher incidence, for example, of comorbidities such as COPD, peripheral vascular disease, and previous cerebrovascular disease, as well as up to 30% urgent cases versus your 12%. I suppose one of the things that I guess I would ask you is, how realistic is this, as I am sure you are evolving more and more into a sicker group of patients? Many of the patients that probably benefit the most from this are the group of patients in the high comorbidity group. So I would ask you, have you further broken down a way of identifying which patients are most likely to be able to be discharged early, and of those that are discharged on that first postoperative day, which are least likely to be readmitted? I think if you can tease that out, that will help all of us.

In addition, I think most studies have demonstrated that to get people out of the hospital early, you need to have a very short time on the ventilator. Do you have any particular anesthetic techniques or tricks or postoperative care techniques in order to mobilize these people and allow them to be discharged early? Thank you.

DR KIT V. AROM (Minneapolis, MN): Clint, I enjoyed your paper very much. Last year we looked at our own results of close to 1,000 off-pump cases. We did not have enough early discharge people like you did, so we are unable to come up with predictors of early discharge. We, however, have predictors of prolonged or delayed discharge, which includes older age, female gender, history of renal failure, and history of stroke.

With your patient population presented here, 56% of them were discharged on day one; therefore, you should be able to report the predictors of early discharge in this group. Thank you.

DR BAISDEN: I want to thank Drs Grover and Arom for their kind comments and interesting questions. We have not analyzed our data to know the predictors of successful discharge, but will do so and incorporate that into the final manuscript for publication.

Dr Grover asked what things we do to make it so that the patients are ready for early discharge. We use short anesthetics and that sort of thing. Dr Randy Bolton, who was my partner during this time and is now the Chief of Cardiac Surgery at the Self Memorial Hospital in Greenwood, South Carolina, worked very hard on this project. He took it on as a personal project to try to see what things he could do to decrease the amount of morbidity the patients suffered and would allow early discharge. Dr Mack’s earlier talk recommended infiltration of the parasternal area with Marcaine, using intravenous Toradol in the operating room, short-acting anesthetics, good anesthesiologists, keeping the patients warm by keeping the operating room warm and the fluids we give the patients warm; all of those things seem to have a positive effect.

Probably the most helpful thing, at least in my observation, was that the patients that looked like they were ready to go home the day after surgery, and who wanted to go home, were the ones who were most successful. I think this has a lot to say about surgeons’ intuition. But we will see if we can get more data in the future. Thank you very much for this opportunity.

	References

Top Abstract Introduction Material and methods Results Comment Discussion References

 

1. Roberts A.J., Woodhall D.D., Conti R.R., et al. Mortality, morbidity, and cost accounting related to coronary artery bypass graft surgery in the elderly. Ann Thorac Surg 1985;39:426-432.[Abstract]
2. Wagner D.P., Wineland T.D., Knaus W.A. The hidden costs of treating severely ill patients: charges and resource consumption in an intensive care unit. Health Care Rev Fin Fall 1983;5:81-86.
3. Krohn B.G., Kay J.H., Mendez M.A., et al. Rapid sustained recovery after cardiac operations. J Thorac Cardiovasc Surg 1990;100:194-197.[Abstract]
4. Ott R.A., Gutfinger D.E., Miller M.P., et al. Coronary artery bypass grafting "on pump": role of three-day discharge. Ann Thorac Surg 1997;64:478-481.[Abstract/Free Full Text]
5. Puskas J.D., Thourani V.H., Marshall J.J., et al. Clinical outcomes, angiographic patency, and resource utilization in 200 consecutive off-pump coronary bypass patients. Ann Thorac Surg 2001;71:1477-1484.[Abstract/Free Full Text]
6. Society of Thoracic Surgeons. Data analyses of the Society of Thoracic Surgeons National Adult Cardiac Surgery Database. Chicago, IL: Society of Thoracic Surgeons, 2001:39.

This article has been cited by other articles:

				A. R. Srivastava, A. Banerjee, D. K. Tempe, B. Mishra, V. Muppiri, S. Narang, H. Singh, and S. Virmani A comprehensive approach to fast tracking in cardiac surgery: ambulatory low-risk open-heart surgery Eur. J. Cardiothorac. Surg., June 1, 2008; 33(6): 955 - 960. [Abstract] [Full Text] [PDF]

				J.W. R. Bolton Same Day Discharge Following Multivessel Off-Pump Coronary Artery Bypass Via Sternotomy Ann. Thorac. Surg., January 1, 2005; 79(1): 345 - 346. [Abstract] [Full Text] [PDF]

				J.W. R. Bolton Ambulatory cardiac surgery Eur. J. Cardiothorac. Surg., December 1, 2004; 26(Suppl_1): S68 - S71. [Abstract] [Full Text] [PDF]

				G. Varela, J. L. Aranda, M. F. Jimenez, and N. Novoa Emergency hospital readmission after major lung resection: prevalence and related variables Eur. J. Cardiothorac. Surg., September 1, 2004; 26(3): 494 - 497. [Abstract] [Full Text] [PDF]

				D. L. Ngaage Off-pump coronary artery bypass grafting: the myth, the logic and the science Eur. J. Cardiothorac. Surg., October 1, 2003; 24(4): 557 - 570. [Abstract] [Full Text] [PDF]

This Article Abstract Full Text (PDF) Alert me when this article is cited Alert me if a correction is posted Citation Map Services Email this article to a friend Similar articles in this journal Similar articles in PubMed Alert me to new issues of the journal Add to Personal Folders Download to citation manager Author home page(s): Clinton E. Baisden Permission Requests Citing Articles Citing Articles via HighWire Citing Articles via Google Scholar Google Scholar Articles by Baisden, C. E. Articles by Riggs, M. W. Search for Related Content PubMed PubMed Citation Articles by Baisden, C. E. Articles by Riggs, M. W. Related Collections Coronary disease